# Claude Docs
## Docs
- [API-Schlüssel abrufen](https://docs.claude.com/de/api/admin-api/apikeys/get-api-key.md)
- [API-Schlüssel auflisten](https://docs.claude.com/de/api/admin-api/apikeys/list-api-keys.md)
- [API-Schlüssel aktualisieren](https://docs.claude.com/de/api/admin-api/apikeys/update-api-key.md)
- [Claude Code Nutzungsbericht abrufen](https://docs.claude.com/de/api/admin-api/claude-code/get-claude-code-usage-report.md): Retrieve daily aggregated usage metrics for Claude Code users.
Enables organizations to analyze developer productivity and build custom dashboards.
- [Einladung erstellen](https://docs.claude.com/de/api/admin-api/invites/create-invite.md)
- [Einladung löschen](https://docs.claude.com/de/api/admin-api/invites/delete-invite.md)
- [Einladung abrufen](https://docs.claude.com/de/api/admin-api/invites/get-invite.md)
- [Einladungen auflisten](https://docs.claude.com/de/api/admin-api/invites/list-invites.md)
- [Organisationsinformationen abrufen](https://docs.claude.com/de/api/admin-api/organization/get-me.md)
- [Kostenbericht abrufen](https://docs.claude.com/de/api/admin-api/usage-cost/get-cost-report.md)
- [Nutzungsbericht für die Messages API abrufen](https://docs.claude.com/de/api/admin-api/usage-cost/get-messages-usage-report.md)
- [Benutzer abrufen](https://docs.claude.com/de/api/admin-api/users/get-user.md)
- [Benutzer auflisten](https://docs.claude.com/de/api/admin-api/users/list-users.md)
- [Benutzer entfernen](https://docs.claude.com/de/api/admin-api/users/remove-user.md)
- [Benutzer aktualisieren](https://docs.claude.com/de/api/admin-api/users/update-user.md)
- [Arbeitsbereich-Mitglied hinzufügen](https://docs.claude.com/de/api/admin-api/workspace_members/create-workspace-member.md)
- [Workspace-Mitglied löschen](https://docs.claude.com/de/api/admin-api/workspace_members/delete-workspace-member.md)
- [Workspace-Mitglied abrufen](https://docs.claude.com/de/api/admin-api/workspace_members/get-workspace-member.md)
- [Arbeitsbereich-Mitglieder auflisten](https://docs.claude.com/de/api/admin-api/workspace_members/list-workspace-members.md)
- [Workspace-Mitglied aktualisieren](https://docs.claude.com/de/api/admin-api/workspace_members/update-workspace-member.md)
- [Arbeitsbereich archivieren](https://docs.claude.com/de/api/admin-api/workspaces/archive-workspace.md)
- [Arbeitsbereich erstellen](https://docs.claude.com/de/api/admin-api/workspaces/create-workspace.md)
- [Arbeitsbereich abrufen](https://docs.claude.com/de/api/admin-api/workspaces/get-workspace.md)
- [Arbeitsbereiche auflisten](https://docs.claude.com/de/api/admin-api/workspaces/list-workspaces.md)
- [Arbeitsbereich aktualisieren](https://docs.claude.com/de/api/admin-api/workspaces/update-workspace.md)
- [Admin API Übersicht](https://docs.claude.com/de/api/administration-api.md)
- [Beta-Header](https://docs.claude.com/de/api/beta-headers.md): Dokumentation zur Verwendung von Beta-Headern mit der Claude API
- [Einen Nachrichten-Batch abbrechen](https://docs.claude.com/de/api/canceling-message-batches.md): Batches may be canceled any time before processing ends. Once cancellation is initiated, the batch enters a `canceling` state, at which time the system may complete any in-progress, non-interruptible requests before finalizing cancellation.
The number of canceled requests is specified in `request_counts`. To determine which requests were canceled, check the individual results within the batch. Note that cancellation may not result in any canceled requests if they were non-interruptible.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Claude Code Analytics API](https://docs.claude.com/de/api/claude-code-analytics-api.md): Greifen Sie programmatisch auf die Claude Code-Nutzungsanalysen und Produktivitätsmetriken Ihrer Organisation mit der Claude Code Analytics Admin API zu.
- [Amazon Bedrock API](https://docs.claude.com/de/api/claude-on-amazon-bedrock.md): Anthropics Claude-Modelle sind jetzt allgemein über Amazon Bedrock verfügbar.
- [Vertex AI API](https://docs.claude.com/de/api/claude-on-vertex-ai.md): Anthropics Claude-Modelle sind jetzt allgemein verfügbar über [Vertex AI](https://cloud.google.com/vertex-ai).
- [Client SDKs](https://docs.claude.com/de/api/client-sdks.md): Wir stellen Client-Bibliotheken in einer Reihe beliebter Sprachen zur Verfügung, die es einfacher machen, mit der Claude API zu arbeiten.
- [Erstelle einen Nachrichten-Batch](https://docs.claude.com/de/api/creating-message-batches.md): Send a batch of Message creation requests.
The Message Batches API can be used to process multiple Messages API requests at once. Once a Message Batch is created, it begins processing immediately. Batches can take up to 24 hours to complete.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Einen Nachrichten-Batch löschen](https://docs.claude.com/de/api/deleting-message-batches.md): Delete a Message Batch.
Message Batches can only be deleted once they've finished processing. If you'd like to delete an in-progress batch, you must first cancel it.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Fehler](https://docs.claude.com/de/api/errors.md)
- [Eine Datei herunterladen](https://docs.claude.com/de/api/files-content.md): Laden Sie den Inhalt einer von Claude generierten Datei herunter
- [Eine Datei erstellen](https://docs.claude.com/de/api/files-create.md): Eine Datei hochladen
- [Eine Datei löschen](https://docs.claude.com/de/api/files-delete.md): Eine Datei über die API unzugänglich machen
- [Dateien auflisten](https://docs.claude.com/de/api/files-list.md): Dateien innerhalb eines Arbeitsbereichs auflisten
- [Datei-Metadaten abrufen](https://docs.claude.com/de/api/files-metadata.md)
- [Hilfe erhalten](https://docs.claude.com/de/api/getting-help.md): Wir haben versucht, die Antworten auf die häufigsten Fragen in dieser Dokumentation bereitzustellen. Wenn Sie jedoch weitere technische Unterstützung bei der Verwendung von Claude, der Claude API oder einem unserer Produkte benötigen, können Sie unser Support-Team unter [support.claude.com](https://support.claude.com) erreichen.
- [Umgang mit Stop-Gründen](https://docs.claude.com/de/api/handling-stop-reasons.md): Verstehen Sie, warum Claude die Antwortgenerierung stoppt und wie Sie verschiedene stop_reason-Werte in Ihren Anwendungen handhaben.
- [IP-Adressen](https://docs.claude.com/de/api/ip-addresses.md): Anthropic-Dienste verwenden feste IP-Adressen sowohl für eingehende als auch für ausgehende Verbindungen. Sie können diese Adressen verwenden, um Ihre Firewall-Regeln für den sicheren Zugriff auf die Claude API und Konsole zu konfigurieren. Diese Adressen werden sich nicht ohne Vorankündigung ändern.
- [Nachrichtenstapel auflisten](https://docs.claude.com/de/api/listing-message-batches.md): List all Message Batches within a Workspace. Most recently created batches are returned first.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Nachrichten](https://docs.claude.com/de/api/messages.md): Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless multi-turn conversations.
Learn more about the Messages API in our [user guide](/en/docs/initial-setup)
- [Message Batches Beispiele](https://docs.claude.com/de/api/messages-batch-examples.md): Beispielverwendung für die Message Batches API
- [Nachrichten-Token zählen](https://docs.claude.com/de/api/messages-count-tokens.md): Count the number of tokens in a Message.
The Token Count API can be used to count the number of tokens in a Message, including tools, images, and documents, without creating it.
Learn more about token counting in our [user guide](/en/docs/build-with-claude/token-counting)
- [Messages Beispiele](https://docs.claude.com/de/api/messages-examples.md): Anfrage- und Antwortbeispiele für die Messages API
- [Migration von Text Completions](https://docs.claude.com/de/api/migrating-from-text-completions-to-messages.md): Migration von Text Completions zu Messages
- [Ein Modell abrufen](https://docs.claude.com/de/api/models.md): Get a specific model.
The Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.
- [Modelle auflisten](https://docs.claude.com/de/api/models-list.md): List available models.
The Models API response can be used to determine which models are available for use in the API. More recently released models are listed first.
- [OpenAI SDK Kompatibilität](https://docs.claude.com/de/api/openai-sdk.md): Anthropic bietet eine Kompatibilitätsschicht, die es Ihnen ermöglicht, das OpenAI SDK zu verwenden, um die Claude API zu testen. Mit wenigen Code-Änderungen können Sie schnell die Fähigkeiten der Anthropic-Modelle bewerten.
- [Übersicht](https://docs.claude.com/de/api/overview.md): Übersicht über die Claude API
- [Einen Prompt generieren](https://docs.claude.com/de/api/prompt-tools-generate.md): Einen gut geschriebenen Prompt generieren
- [Einen Prompt verbessern](https://docs.claude.com/de/api/prompt-tools-improve.md): Erstellen Sie einen neuen und verbesserten Prompt, der durch Feedback geleitet wird
- [Einen Prompt als Vorlage erstellen](https://docs.claude.com/de/api/prompt-tools-templatize.md): Einen Prompt als Vorlage erstellen, indem Variablen identifiziert und extrahiert werden
- [Ratenlimits](https://docs.claude.com/de/api/rate-limits.md): Um Missbrauch zu verhindern und die Kapazität unserer API zu verwalten, haben wir Limits implementiert, wie viel eine Organisation die Claude API nutzen kann.
- [Nachrichtenstapel-Ergebnisse abrufen](https://docs.claude.com/de/api/retrieving-message-batch-results.md): Streams the results of a Message Batch as a `.jsonl` file.
Each line in the file is a JSON object containing the result of a single request in the Message Batch. Results are not guaranteed to be in the same order as requests. Use the `custom_id` field to match results to requests.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Eine Nachrichten-Batch abrufen](https://docs.claude.com/de/api/retrieving-message-batches.md): This endpoint is idempotent and can be used to poll for Message Batch completion. To access the results of a Message Batch, make a request to the `results_url` field in the response.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Service-Stufen](https://docs.claude.com/de/api/service-tiers.md): Verschiedene Service-Stufen ermöglichen es Ihnen, Verfügbarkeit, Leistung und vorhersagbare Kosten basierend auf den Bedürfnissen Ihrer Anwendung auszubalancieren.
- [Unterstützte Regionen](https://docs.claude.com/de/api/supported-regions.md): Hier sind die Länder, Regionen und Gebiete, für die wir derzeit den Zugang unterstützen können:
- [Nutzungs- und Kosten-API](https://docs.claude.com/de/api/usage-cost-api.md): Greifen Sie programmatisch auf die API-Nutzungs- und Kostendaten Ihrer Organisation mit der Nutzungs- und Kosten-Admin-API zu.
- [Versionen](https://docs.claude.com/de/api/versioning.md): Beim Stellen von API-Anfragen müssen Sie einen `anthropic-version` Request-Header senden. Zum Beispiel `anthropic-version: 2023-06-01`. Wenn Sie unsere [Client-SDKs](/de/api/client-sdks) verwenden, wird dies automatisch für Sie erledigt.
- [Glossar](https://docs.claude.com/de/docs/about-claude/glossary.md): Diese Konzepte sind nicht einzigartig für Anthropics Sprachmodelle, aber wir präsentieren unten eine kurze Zusammenfassung der wichtigsten Begriffe.
- [Modell-Veraltungen](https://docs.claude.com/de/docs/about-claude/model-deprecations.md)
- [Das richtige Modell auswählen](https://docs.claude.com/de/docs/about-claude/models/choosing-a-model.md): Die Auswahl des optimalen Claude-Modells für Ihre Anwendung erfordert die Abwägung von drei wichtigen Faktoren: Fähigkeiten, Geschwindigkeit und Kosten. Dieser Leitfaden hilft Ihnen, eine fundierte Entscheidung auf Basis Ihrer spezifischen Anforderungen zu treffen.
- [Migration zu Claude 4](https://docs.claude.com/de/docs/about-claude/models/migrating-to-claude-4.md): Leitfaden für die Migration von Claude 3.7 Modellen zu Claude 4 Modellen (Opus 4 und Sonnet 4).
- [Modelle Übersicht](https://docs.claude.com/de/docs/about-claude/models/overview.md): Claude ist eine Familie von hochmodernen großen Sprachmodellen, die von Anthropic entwickelt wurden. Dieser Leitfaden stellt unsere Modelle vor und vergleicht ihre Leistung mit älteren Modellen.
- [Preise](https://docs.claude.com/de/docs/about-claude/pricing.md): Erfahren Sie mehr über Anthropics Preisstruktur für Modelle und Funktionen
- [Content-Moderation](https://docs.claude.com/de/docs/about-claude/use-case-guides/content-moderation.md): Content-Moderation ist ein kritischer Aspekt für die Aufrechterhaltung einer sicheren, respektvollen und produktiven Umgebung in digitalen Anwendungen. In diesem Leitfaden besprechen wir, wie Claude zur Moderation von Inhalten in Ihrer digitalen Anwendung verwendet werden kann.
- [Kundensupport-Agent](https://docs.claude.com/de/docs/about-claude/use-case-guides/customer-support-chat.md): Dieser Leitfaden zeigt, wie Sie Claudes fortgeschrittene Konversationsfähigkeiten nutzen können, um Kundenanfragen in Echtzeit zu bearbeiten, 24/7-Support zu bieten, Wartezeiten zu reduzieren und hohe Support-Volumina mit präzisen Antworten und positiven Interaktionen zu bewältigen.
- [Rechtliche Zusammenfassung](https://docs.claude.com/de/docs/about-claude/use-case-guides/legal-summarization.md): Dieser Leitfaden führt Sie durch die Nutzung von Claudes fortschrittlichen Fähigkeiten zur Verarbeitung natürlicher Sprache, um rechtliche Dokumente effizient zusammenzufassen, wichtige Informationen zu extrahieren und die rechtliche Recherche zu beschleunigen. Mit Claude können Sie die Überprüfung von Verträgen, die Vorbereitung von Rechtsstreitigkeiten und regulatorische Arbeiten rationalisieren, Zeit sparen und die Genauigkeit in Ihren rechtlichen Prozessen sicherstellen.
- [Leitfäden für häufige Anwendungsfälle](https://docs.claude.com/de/docs/about-claude/use-case-guides/overview.md)
- [Ticket-Routing](https://docs.claude.com/de/docs/about-claude/use-case-guides/ticket-routing.md): Dieser Leitfaden führt durch die Nutzung von Claudes fortschrittlichen Fähigkeiten zum Verstehen natürlicher Sprache, um Kundensupport-Tickets skaliert basierend auf Kundenabsicht, Dringlichkeit, Priorisierung, Kundenprofil und mehr zu klassifizieren.
- [Google Sheets Add-on](https://docs.claude.com/de/docs/agents-and-tools/claude-for-sheets.md): Die [Claude for Sheets Erweiterung](https://workspace.google.com/marketplace/app/claude%5Ffor%5Fsheets/909417792257) integriert Claude in Google Sheets und ermöglicht es Ihnen, Interaktionen mit Claude direkt in Zellen auszuführen.
- [MCP-Connector](https://docs.claude.com/de/docs/agents-and-tools/mcp-connector.md): Verbinden Sie sich mit entfernten MCP-Servern direkt über die Messages API ohne separaten MCP-Client.
- [Remote MCP-Server](https://docs.claude.com/de/docs/agents-and-tools/remote-mcp-servers.md): Remote MCP-Server
- [Bash-Tool](https://docs.claude.com/de/docs/agents-and-tools/tool-use/bash-tool.md): Das Bash-Tool ermöglicht es Claude, Shell-Befehle in einer persistenten Bash-Sitzung auszuführen und dabei Systemoperationen, Skriptausführung und Kommandozeilen-Automatisierung zu ermöglichen.
- [Code-Ausführungs-Tool](https://docs.claude.com/de/docs/agents-and-tools/tool-use/code-execution-tool.md)
- [Computer-Nutzungs-Tool](https://docs.claude.com/de/docs/agents-and-tools/tool-use/computer-use-tool.md): Claude kann mit Computerumgebungen über das Computer-Nutzungs-Tool interagieren, das Screenshot-Funktionen und Maus-/Tastatursteuerung für autonome Desktop-Interaktion bietet.
- [Feinkörniges Tool-Streaming](https://docs.claude.com/de/docs/agents-and-tools/tool-use/fine-grained-tool-streaming.md): Tool-Verwendung unterstützt jetzt feinkörniges Streaming für Parameterwerte, wodurch Entwickler Tool-Use-Parameter ohne Pufferung/JSON-Validierung streamen können.
- [Wie man Tool-Nutzung implementiert](https://docs.claude.com/de/docs/agents-and-tools/tool-use/implement-tool-use.md): Eine umfassende Anleitung zur Implementierung von Tool-Nutzung mit Claude, einschließlich Modellauswahl, Tool-Definitionen, Ausgabekontrolle und Fehlerbehandlung.
- [Tool-Nutzung mit Claude](https://docs.claude.com/de/docs/agents-and-tools/tool-use/overview.md)
- [Text-Editor-Tool](https://docs.claude.com/de/docs/agents-and-tools/tool-use/text-editor-tool.md): Claude kann ein von Anthropic definiertes Text-Editor-Tool verwenden, um Textdateien anzuzeigen und zu bearbeiten, und Ihnen dabei helfen, Ihren Code oder andere Textdokumente zu debuggen, zu reparieren und zu verbessern.
- [Token-effiziente Tool-Nutzung](https://docs.claude.com/de/docs/agents-and-tools/tool-use/token-efficient-tool-use.md)
- [Web-Abruf-Tool](https://docs.claude.com/de/docs/agents-and-tools/tool-use/web-fetch-tool.md): Das Web-Abruf-Tool ermöglicht es Claude, vollständige Inhalte von angegebenen Webseiten und PDF-Dokumenten abzurufen.
- [Web-Suche-Tool](https://docs.claude.com/de/docs/agents-and-tools/tool-use/web-search-tool.md): Das Web-Suche-Tool gibt Claude direkten Zugang zu Echtzeit-Webinhalten und ermöglicht es, Fragen mit aktuellen Informationen zu beantworten.
- [Batch-Verarbeitung](https://docs.claude.com/de/docs/build-with-claude/batch-processing.md)
- [Zitate](https://docs.claude.com/de/docs/build-with-claude/citations.md): Claude ist in der Lage, detaillierte Zitate zu liefern, wenn es Fragen zu Dokumenten beantwortet, und hilft Ihnen dabei, Informationsquellen in Antworten zu verfolgen und zu verifizieren.
- [Kontextfenster](https://docs.claude.com/de/docs/build-with-claude/context-windows.md)
- [Embeddings](https://docs.claude.com/de/docs/build-with-claude/embeddings.md): Text-Embeddings sind numerische Darstellungen von Text, die es ermöglichen, semantische Ähnlichkeit zu messen. Dieser Leitfaden führt in Embeddings, ihre Anwendungen und die Verwendung von Embedding-Modellen für Aufgaben wie Suche, Empfehlungen und Anomalieerkennung ein.
- [Entwickeln mit erweitertem Denken](https://docs.claude.com/de/docs/build-with-claude/extended-thinking.md)
- [Files API](https://docs.claude.com/de/docs/build-with-claude/files.md): Die Files API ermöglicht es Ihnen, Dateien hochzuladen und zu verwalten, um sie mit der Claude API zu verwenden, ohne Inhalte bei jeder Anfrage erneut hochladen zu müssen.
- [Mehrsprachige Unterstützung](https://docs.claude.com/de/docs/build-with-claude/multilingual-support.md): Claude überzeugt bei Aufgaben in mehreren Sprachen und behält eine starke sprachübergreifende Leistung im Vergleich zum Englischen bei.
- [Funktionsübersicht](https://docs.claude.com/de/docs/build-with-claude/overview.md): Entdecken Sie Claudes erweiterte Funktionen und Fähigkeiten.
- [PDF-Unterstützung](https://docs.claude.com/de/docs/build-with-claude/pdf-support.md): Verarbeiten Sie PDFs mit Claude. Extrahieren Sie Text, analysieren Sie Diagramme und verstehen Sie visuelle Inhalte aus Ihren Dokumenten.
- [Prompt-Caching](https://docs.claude.com/de/docs/build-with-claude/prompt-caching.md)
- [Sei klar, direkt und detailliert](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/be-clear-and-direct.md)
- [Claude denken lassen (Chain of Thought Prompting) zur Leistungssteigerung](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/chain-of-thought.md)
- [Komplexe Prompts für bessere Leistung verketten](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/chain-prompts.md)
- [Claude 4 Prompt Engineering Best Practices](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/claude-4-best-practices.md): Spezifische Prompt-Engineering-Techniken für Claude 4-Modelle zur Erzielung optimaler Ergebnisse in Ihren Anwendungen.
- [Tipps für erweiterte Denkprozesse](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/extended-thinking-tips.md)
- [Tipps für Prompting mit langem Kontext](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/long-context-tips.md)
- [Beispiele verwenden (Multishot-Prompting) um Claudes Verhalten zu steuern](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/multishot-prompting.md): Lernen Sie, wie Sie mit Beispielen (Multishot-Prompting) die Genauigkeit und Konsistenz von Claudes Ausgaben verbessern können.
- [Prompt Engineering Übersicht](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/overview.md)
- [Claudes Antwort vorfüllen für bessere Ausgabekontrolle](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/prefill-claudes-response.md)
- [Automatische Generierung von Prompt-Vorlagen im ersten Entwurf](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/prompt-generator.md): Ein Tool zur automatischen Generierung von Prompt-Vorlagen für Claude
- [Verwenden Sie unseren Prompt-Verbesserer, um Ihre Prompts zu optimieren](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/prompt-improver.md)
- [Prompt-Vorlagen und Variablen verwenden](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/prompt-templates-and-variables.md)
- [Claude eine Rolle mit einem System-Prompt zuweisen](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/system-prompts.md)
- [XML-Tags zur Strukturierung Ihrer Prompts verwenden](https://docs.claude.com/de/docs/build-with-claude/prompt-engineering/use-xml-tags.md)
- [Suchergebnisse](https://docs.claude.com/de/docs/build-with-claude/search-results.md): Ermöglichen Sie natürliche Zitate für RAG-Anwendungen durch die Bereitstellung von Suchergebnissen mit Quellenangabe
- [Streaming-Nachrichten](https://docs.claude.com/de/docs/build-with-claude/streaming.md): Erfahren Sie, wie Sie Nachrichten mit Server-Sent Events (SSE) inkrementell streamen können.
- [Token-Zählung](https://docs.claude.com/de/docs/build-with-claude/token-counting.md)
- [Vision](https://docs.claude.com/de/docs/build-with-claude/vision.md): Die Claude 3 und 4 Modellfamilien verfügen über neue Vision-Fähigkeiten, die es Claude ermöglichen, Bilder zu verstehen und zu analysieren, was aufregende Möglichkeiten für multimodale Interaktionen eröffnet.
- [Claude Code auf Amazon Bedrock](https://docs.claude.com/de/docs/claude-code/amazon-bedrock.md): Erfahren Sie mehr über die Konfiguration von Claude Code über Amazon Bedrock, einschließlich Setup, IAM-Konfiguration und Fehlerbehebung.
- [Analytics](https://docs.claude.com/de/docs/claude-code/analytics.md): Zeigen Sie detaillierte Nutzungseinblicke und Produktivitätsmetriken für die Claude Code-Bereitstellung Ihrer Organisation an.
- [CLI-Referenz](https://docs.claude.com/de/docs/claude-code/cli-reference.md): Vollständige Referenz für die Claude Code Kommandozeilen-Schnittstelle, einschließlich Befehle und Flags.
- [Häufige Arbeitsabläufe](https://docs.claude.com/de/docs/claude-code/common-workflows.md): Lernen Sie häufige Arbeitsabläufe mit Claude Code kennen.
- [Kosten effektiv verwalten](https://docs.claude.com/de/docs/claude-code/costs.md): Erfahren Sie, wie Sie die Token-Nutzung und Kosten bei der Verwendung von Claude Code verfolgen und optimieren können.
- [Datennutzung](https://docs.claude.com/de/docs/claude-code/data-usage.md): Erfahren Sie mehr über Anthropics Datennutzungsrichtlinien für Claude
- [Entwicklungscontainer](https://docs.claude.com/de/docs/claude-code/devcontainer.md): Erfahren Sie mehr über den Claude Code-Entwicklungscontainer für Teams, die konsistente, sichere Umgebungen benötigen.
- [Claude Code GitHub Actions](https://docs.claude.com/de/docs/claude-code/github-actions.md): Erfahren Sie, wie Sie Claude Code mit Claude Code GitHub Actions in Ihren Entwicklungsworkflow integrieren
- [Claude Code GitLab CI/CD](https://docs.claude.com/de/docs/claude-code/gitlab-ci-cd.md): Erfahren Sie, wie Sie Claude Code in Ihren Entwicklungsworkflow mit GitLab CI/CD integrieren
- [Claude Code auf Google Vertex AI](https://docs.claude.com/de/docs/claude-code/google-vertex-ai.md): Erfahren Sie mehr über die Konfiguration von Claude Code über Google Vertex AI, einschließlich Setup, IAM-Konfiguration und Fehlerbehebung.
- [Hooks-Referenz](https://docs.claude.com/de/docs/claude-code/hooks.md): Diese Seite bietet Referenzdokumentation für die Implementierung von Hooks in Claude Code.
- [Erste Schritte mit Claude Code Hooks](https://docs.claude.com/de/docs/claude-code/hooks-guide.md): Erfahren Sie, wie Sie das Verhalten von Claude Code durch die Registrierung von Shell-Befehlen anpassen und erweitern können
- [Identitäts- und Zugriffsverwaltung](https://docs.claude.com/de/docs/claude-code/iam.md): Erfahren Sie, wie Sie Benutzerauthentifizierung, Autorisierung und Zugriffskontrollen für Claude Code in Ihrer Organisation konfigurieren.
- [Claude Code zu Ihrer IDE hinzufügen](https://docs.claude.com/de/docs/claude-code/ide-integrations.md): Erfahren Sie, wie Sie Claude Code zu Ihrer bevorzugten IDE hinzufügen
- [Interaktiver Modus](https://docs.claude.com/de/docs/claude-code/interactive-mode.md): Vollständige Referenz für Tastenkombinationen, Eingabemodi und interaktive Funktionen in Claude Code-Sitzungen.
- [Rechtliches und Compliance](https://docs.claude.com/de/docs/claude-code/legal-and-compliance.md): Rechtliche Vereinbarungen, Compliance-Zertifizierungen und Sicherheitsinformationen für Claude Code.
- [LLM-Gateway-Konfiguration](https://docs.claude.com/de/docs/claude-code/llm-gateway.md): Erfahren Sie, wie Sie Claude Code mit LLM-Gateway-Lösungen konfigurieren, einschließlich LiteLLM-Setup, Authentifizierungsmethoden und Unternehmensfunktionen wie Nutzungsverfolgung und Budgetverwaltung.
- [Claude Code über MCP mit Tools verbinden](https://docs.claude.com/de/docs/claude-code/mcp.md): Erfahren Sie, wie Sie Claude Code über das Model Context Protocol mit Ihren Tools verbinden.
- [Claudes Speicher verwalten](https://docs.claude.com/de/docs/claude-code/memory.md): Erfahren Sie, wie Sie den Speicher von Claude Code sitzungsübergreifend mit verschiedenen Speicherorten und bewährten Praktiken verwalten.
- [Modellkonfiguration](https://docs.claude.com/de/docs/claude-code/model-config.md): Erfahren Sie mehr über die Claude Code Modellkonfiguration, einschließlich Modellaliase wie `opusplan`
- [Überwachung](https://docs.claude.com/de/docs/claude-code/monitoring-usage.md): Erfahren Sie, wie Sie OpenTelemetry für Claude Code aktivieren und konfigurieren.
- [Enterprise-Netzwerkkonfiguration](https://docs.claude.com/de/docs/claude-code/network-config.md): Konfigurieren Sie Claude Code für Unternehmensumgebungen mit Proxy-Servern, benutzerdefinierten Certificate Authorities (CA) und gegenseitiger Transport Layer Security (mTLS) Authentifizierung.
- [Ausgabestile](https://docs.claude.com/de/docs/claude-code/output-styles.md): Claude Code für Anwendungen jenseits der Softwareentwicklung anpassen
- [Claude Code Übersicht](https://docs.claude.com/de/docs/claude-code/overview.md): Erfahren Sie mehr über Claude Code, Anthropics agentisches Coding-Tool, das in Ihrem Terminal lebt und Ihnen hilft, Ideen schneller denn je in Code zu verwandeln.
- [Schnellstart](https://docs.claude.com/de/docs/claude-code/quickstart.md): Willkommen bei Claude Code!
- [Benutzerdefinierte Tools](https://docs.claude.com/de/docs/claude-code/sdk/custom-tools.md): Erstellen und integrieren Sie benutzerdefinierte Tools, um die Funktionalität des Claude Code SDK zu erweitern
- [System-Prompts modifizieren](https://docs.claude.com/de/docs/claude-code/sdk/modifying-system-prompts.md): Lernen Sie, wie Sie Claudes Verhalten durch Modifikation von System-Prompts mit drei Ansätzen anpassen können - Output-Styles, appendSystemPrompt und customSystemPrompt.
- [Kosten und Nutzung verfolgen](https://docs.claude.com/de/docs/claude-code/sdk/sdk-cost-tracking.md): Token-Nutzung für die Abrechnung im Claude Code SDK verstehen und verfolgen
- [Headless-Modus](https://docs.claude.com/de/docs/claude-code/sdk/sdk-headless.md): Claude Code programmatisch ohne interaktive Benutzeroberfläche ausführen
- [MCP im SDK](https://docs.claude.com/de/docs/claude-code/sdk/sdk-mcp.md): Erweitern Sie Claude Code mit benutzerdefinierten Tools mithilfe von Model Context Protocol-Servern
- [Übersicht](https://docs.claude.com/de/docs/claude-code/sdk/sdk-overview.md): Erstellen Sie benutzerdefinierte KI-Agenten mit dem Claude Code SDK
- [Berechtigungen verwalten](https://docs.claude.com/de/docs/claude-code/sdk/sdk-permissions.md): Steuern Sie die Tool-Nutzung und Berechtigungen im Claude Code SDK
- [Python SDK Referenz](https://docs.claude.com/de/docs/claude-code/sdk/sdk-python.md): Vollständige API-Referenz für das Claude Code Python SDK, einschließlich aller Funktionen, Typen und Klassen.
- [Session-Verwaltung](https://docs.claude.com/de/docs/claude-code/sdk/sdk-sessions.md): Verstehen, wie das Claude Code SDK Sessions und Session-Wiederaufnahme handhabt
- [Slash-Befehle im SDK](https://docs.claude.com/de/docs/claude-code/sdk/sdk-slash-commands.md): Lernen Sie, wie Sie Slash-Befehle verwenden, um Claude Code-Sitzungen über das SDK zu steuern
- [TypeScript SDK Referenz](https://docs.claude.com/de/docs/claude-code/sdk/sdk-typescript.md): Vollständige API-Referenz für das Claude Code TypeScript SDK, einschließlich aller Funktionen, Typen und Schnittstellen.
- [Streaming-Eingabe](https://docs.claude.com/de/docs/claude-code/sdk/streaming-vs-single-mode.md): Verstehen der zwei Eingabemodi für das Claude Code SDK und wann jeder verwendet werden sollte
- [Subagenten im SDK](https://docs.claude.com/de/docs/claude-code/sdk/subagents.md): Arbeiten mit Subagenten im Claude Code SDK
- [Todo-Listen](https://docs.claude.com/de/docs/claude-code/sdk/todo-tracking.md): Verfolgen und zeigen Sie Todos mit dem Claude Code SDK für organisiertes Aufgabenmanagement an
- [Sicherheit](https://docs.claude.com/de/docs/claude-code/security.md): Erfahren Sie mehr über Claude Codes Sicherheitsvorkehrungen und bewährte Praktiken für die sichere Nutzung.
- [Claude Code Einstellungen](https://docs.claude.com/de/docs/claude-code/settings.md): Konfigurieren Sie Claude Code mit globalen und projektspezifischen Einstellungen sowie Umgebungsvariablen.
- [Claude Code einrichten](https://docs.claude.com/de/docs/claude-code/setup.md): Installieren, authentifizieren und beginnen Sie mit der Verwendung von Claude Code auf Ihrem Entwicklungsrechner.
- [Slash-Befehle](https://docs.claude.com/de/docs/claude-code/slash-commands.md): Steuern Sie Claudes Verhalten während einer interaktiven Sitzung mit Slash-Befehlen.
- [Statuszeilen-Konfiguration](https://docs.claude.com/de/docs/claude-code/statusline.md): Erstellen Sie eine benutzerdefinierte Statuszeile für Claude Code zur Anzeige kontextueller Informationen
- [Subagenten](https://docs.claude.com/de/docs/claude-code/sub-agents.md): Erstellen und verwenden Sie spezialisierte KI-Subagenten in Claude Code für aufgabenspezifische Arbeitsabläufe und verbesserte Kontextverwaltung.
- [Optimieren Sie Ihre Terminal-Einrichtung](https://docs.claude.com/de/docs/claude-code/terminal-config.md): Claude Code funktioniert am besten, wenn Ihr Terminal ordnungsgemäß konfiguriert ist. Befolgen Sie diese Richtlinien, um Ihre Erfahrung zu optimieren.
- [Übersicht über Enterprise-Bereitstellung](https://docs.claude.com/de/docs/claude-code/third-party-integrations.md): Erfahren Sie, wie Claude Code mit verschiedenen Drittanbieterdiensten und Infrastrukturen integriert werden kann, um Enterprise-Bereitstellungsanforderungen zu erfüllen.
- [Fehlerbehebung](https://docs.claude.com/de/docs/claude-code/troubleshooting.md): Entdecken Sie Lösungen für häufige Probleme bei der Installation und Verwendung von Claude Code.
- [Erste Schritte mit Claude](https://docs.claude.com/de/docs/get-started.md): Führen Sie Ihren ersten API-Aufruf an Claude durch und erstellen Sie einen einfachen Web-Such-Assistenten
- [Einführung in Claude](https://docs.claude.com/de/docs/intro.md): Claude ist eine hochperformante, vertrauenswürdige und intelligente KI-Plattform von Anthropic. Claude zeichnet sich bei Aufgaben aus, die Sprache, Argumentation, Analyse, Programmierung und mehr umfassen.
- [Anthropic Datenschutzerklärung](https://docs.claude.com/de/docs/legal-center/privacy.md)
- [Sicherheit und Compliance](https://docs.claude.com/de/docs/legal-center/security-compliance.md)
- [Mit Claude entwickeln](https://docs.claude.com/de/docs/overview.md): Claude ist eine Familie von [hochleistungsfähigen und intelligenten KI-Modellen](/de/docs/about-claude/models), die von Anthropic entwickelt wurden. Während Claude mächtig und erweiterbar ist, ist es auch die vertrauenswürdigste und zuverlässigste verfügbare KI. Es folgt kritischen Protokollen, macht weniger Fehler und ist resistent gegen Jailbreaks—was es [Unternehmenskunden](https://claude.com/customers) ermöglicht, die sichersten KI-gestützten Anwendungen im großen Maßstab zu entwickeln.
- [Definieren Sie Ihre Erfolgskriterien](https://docs.claude.com/de/docs/test-and-evaluate/define-success.md)
- [Starke empirische Evaluierungen erstellen](https://docs.claude.com/de/docs/test-and-evaluate/develop-tests.md): Entwickeln Sie Testfälle zur Messung der LLM-Leistung gegen Ihre Erfolgskriterien.
- [Verwendung des Evaluierungs-Tools](https://docs.claude.com/de/docs/test-and-evaluate/eval-tool.md): Die [Claude Console](https://console.anthropic.com/dashboard) verfügt über ein **Evaluierungs-Tool**, mit dem Sie Ihre Prompts unter verschiedenen Szenarien testen können.
- [null](https://docs.claude.com/de/docs/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals.md)
- [Ausgabekonsistenz erhöhen (JSON-Modus)](https://docs.claude.com/de/docs/test-and-evaluate/strengthen-guardrails/increase-consistency.md)
- [Claude mit Rollenprompting und Vorbefüllung im Charakter halten](https://docs.claude.com/de/docs/test-and-evaluate/strengthen-guardrails/keep-claude-in-character.md)
- [Jailbreaks und Prompt-Injections abwehren](https://docs.claude.com/de/docs/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks.md)
- [Halluzinationen reduzieren](https://docs.claude.com/de/docs/test-and-evaluate/strengthen-guardrails/reduce-hallucinations.md)
- [Latenz reduzieren](https://docs.claude.com/de/docs/test-and-evaluate/strengthen-guardrails/reduce-latency.md)
- [Prompt-Leaks reduzieren](https://docs.claude.com/de/docs/test-and-evaluate/strengthen-guardrails/reduce-prompt-leak.md)
- [null](https://docs.claude.com/de/home.md)
- [Claude Entwicklerplattform](https://docs.claude.com/de/release-notes/api.md): Verfolgen Sie Updates zur Claude API und der Claude Console.
- [Claude Apps](https://docs.claude.com/de/release-notes/claude-apps.md): Verfolgen Sie Updates zu Anthropics Claude-Anwendungen.
- [Claude Code](https://docs.claude.com/de/release-notes/claude-code.md): Bleiben Sie über die neuesten Funktionen und Verbesserungen von Claude Code auf dem Laufenden
- [Übersicht](https://docs.claude.com/de/release-notes/overview.md): Verfolgen Sie Updates zu Anthropics Produkten und Dienstleistungen.
- [System Prompts](https://docs.claude.com/de/release-notes/system-prompts.md): Siehe Updates zu den Kern-System-Prompts auf [Claude.ai](https://www.claude.ai) und den Claude [iOS](http://anthropic.com/ios) und [Android](http://anthropic.com/android) Apps.
- [Claude 3.7 Systemkarte](https://docs.claude.com/de/resources/claude-3-7-system-card.md)
- [Claude 3 Modellkarte](https://docs.claude.com/de/resources/claude-3-model-card.md)
- [Claude 4 System Card](https://docs.claude.com/de/resources/claude-4-system-card.md)
- [Anthropic Kurse](https://docs.claude.com/de/resources/courses.md)
- [null](https://docs.claude.com/de/resources/overview.md)
- [Adaptiver Editor](https://docs.claude.com/de/resources/prompt-library/adaptive-editor.md): Text nach benutzerdefinierten Anweisungen umschreiben, z.B. mit einem anderen Ton, Publikum oder Stil.
- [Flughafen-Code-Analyst](https://docs.claude.com/de/resources/prompt-library/airport-code-analyst.md): Finden und extrahieren Sie Flughafen-Codes aus Text.
- [Außerirdischer Anthropologe](https://docs.claude.com/de/resources/prompt-library/alien-anthropologist.md): Analysiere menschliche Kultur und Bräuche aus der Perspektive eines außerirdischen Anthropologen.
- [Alliterations-Alchemist](https://docs.claude.com/de/resources/prompt-library/alliteration-alchemist.md): Generiere alliterative Phrasen und Sätze für jedes gegebene Thema.
- [Babels Sendungen](https://docs.claude.com/de/resources/prompt-library/babels-broadcasts.md): Erstelle überzeugende Produktankündigungs-Tweets in den 10 meistgesprochenen Sprachen der Welt.
- [Markenentwickler](https://docs.claude.com/de/resources/prompt-library/brand-builder.md): Erstelle ein Design-Brief für eine ganzheitliche Markenidentität.
- [Karriere-Coach](https://docs.claude.com/de/resources/prompt-library/career-coach.md): Führen Sie Rollenspiel-Gespräche mit einem KI-Karriere-Coach.
- [Quellen zitieren](https://docs.claude.com/de/resources/prompt-library/cite-your-sources.md): Erhalten Sie Antworten auf Fragen zum Inhalt eines Dokuments mit relevanten Zitaten, die die Antwort unterstützen.
- [Code-Erklärer](https://docs.claude.com/de/resources/prompt-library/code-clarifier.md): Vereinfachen und erklären Sie komplexen Code in einfacher Sprache.
- [Code-Berater](https://docs.claude.com/de/resources/prompt-library/code-consultant.md): Schlagen Sie Verbesserungen zur Optimierung der Python-Code-Performance vor.
- [Unternehmens-Hellseher](https://docs.claude.com/de/resources/prompt-library/corporate-clairvoyant.md): Extrahieren Sie Erkenntnisse, identifizieren Sie Risiken und destillieren Sie wichtige Informationen aus langen Unternehmensberichten in ein einziges Memo.
- [Kosmische Tastenanschläge](https://docs.claude.com/de/resources/prompt-library/cosmic-keystrokes.md): Generiere ein interaktives Schnelltipp-Spiel in einer einzigen HTML-Datei mit seitlichem Scrolling-Gameplay und Tailwind CSS-Styling.
- [CSV-Konverter](https://docs.claude.com/de/resources/prompt-library/csv-converter.md): Konvertieren Sie Daten aus verschiedenen Formaten (JSON, XML, etc.) in ordnungsgemäß formatierte CSV-Dateien.
- [Kulinarischer Schöpfer](https://docs.claude.com/de/resources/prompt-library/culinary-creator.md): Schlagen Sie Rezeptideen basierend auf den verfügbaren Zutaten und Ernährungsvorlieben des Benutzers vor.
- [Datenorganisator](https://docs.claude.com/de/resources/prompt-library/data-organizer.md): Verwandeln Sie unstrukturierten Text in maßgeschneiderte JSON-Tabellen.
- [Anweisungsdecoder](https://docs.claude.com/de/resources/prompt-library/direction-decoder.md): Verwandeln Sie natürliche Sprache in schrittweise Anweisungen.
- [Traumdeuter](https://docs.claude.com/de/resources/prompt-library/dream-interpreter.md): Bieten Sie Interpretationen und Einblicke in die Symbolik der Träume des Benutzers.
- [Effizienz-Schätzer](https://docs.claude.com/de/resources/prompt-library/efficiency-estimator.md): Berechne die Zeitkomplexität von Funktionen und Algorithmen.
- [E-Mail-Extraktor](https://docs.claude.com/de/resources/prompt-library/email-extractor.md): E-Mail-Adressen aus einem Dokument in eine JSON-formatierte Liste extrahieren.
- [Emoji-Encoder](https://docs.claude.com/de/resources/prompt-library/emoji-encoder.md): Wandeln Sie einfachen Text in lustige und ausdrucksstarke Emoji-Nachrichten um.
- [Navigator für ethische Dilemmata](https://docs.claude.com/de/resources/prompt-library/ethical-dilemma-navigator.md): Hilft dem Benutzer dabei, komplexe ethische Dilemmata zu durchdenken und verschiedene Perspektiven zu bieten.
- [Excel-Formel-Experte](https://docs.claude.com/de/resources/prompt-library/excel-formula-expert.md): Erstelle Excel-Formeln basierend auf benutzerbeschriebenen Berechnungen oder Datenmanipulationen.
- [Funktions-Fabrikator](https://docs.claude.com/de/resources/prompt-library/function-fabricator.md): Erstelle Python-Funktionen basierend auf detaillierten Spezifikationen.
- [Futuristischer Modeberater](https://docs.claude.com/de/resources/prompt-library/futuristic-fashion-advisor.md): Schlage avantgardistische Modetrends und -stile für die spezifischen Vorlieben des Benutzers vor.
- [Git gut](https://docs.claude.com/de/resources/prompt-library/git-gud.md): Generiere passende Git-Befehle basierend auf benutzerbeschriebenen Versionskontrollaktionen.
- [Google Apps Scripter](https://docs.claude.com/de/resources/prompt-library/google-apps-scripter.md): Generiere Google Apps Scripts zur Erledigung von Aufgaben basierend auf Benutzeranforderungen.
- [Bewertungsguru](https://docs.claude.com/de/resources/prompt-library/grading-guru.md): Vergleichen und bewerten Sie die Qualität schriftlicher Texte basierend auf benutzerdefinierten Kriterien und Standards.
- [Grammatik-Genie](https://docs.claude.com/de/resources/prompt-library/grammar-genie.md): Verwandeln Sie grammatikalisch falsche Sätze in korrektes Deutsch.
- [Hal der humorvolle Helfer](https://docs.claude.com/de/resources/prompt-library/hal-the-humorous-helper.md): Chatte mit einer sachkundigen KI, die eine sarkastische Seite hat.
- [Redewendungs-Erklärer](https://docs.claude.com/de/resources/prompt-library/idiom-illuminator.md): Erkläre die Bedeutung und den Ursprung gängiger Redewendungen und Sprichwörter.
- [Ersteller von Interviewfragen](https://docs.claude.com/de/resources/prompt-library/interview-question-crafter.md): Generiere Fragen für Interviews.
- [LaTeX-Legende](https://docs.claude.com/de/resources/prompt-library/latex-legend.md): Schreiben Sie LaTeX-Dokumente und generieren Sie Code für mathematische Gleichungen, Tabellen und mehr.
- [Unterrichtsplaner](https://docs.claude.com/de/resources/prompt-library/lesson-planner.md): Erstelle detaillierte Unterrichtspläne zu jedem Thema.
- [Prompt-Bibliothek](https://docs.claude.com/de/resources/prompt-library/library.md)
- [Master-Moderator](https://docs.claude.com/de/resources/prompt-library/master-moderator.md): Bewerte Benutzereingaben auf potenziell schädliche oder illegale Inhalte.
- [Meeting-Protokollant](https://docs.claude.com/de/resources/prompt-library/meeting-scribe.md): Destilliere Meetings zu prägnanten Zusammenfassungen einschließlich Diskussionsthemen, wichtiger Erkenntnisse und Aktionspunkten.
- [Memo-Meister](https://docs.claude.com/de/resources/prompt-library/memo-maestro.md): Verfassen Sie umfassende Unternehmensmemos basierend auf Schlüsselpunkten.
- [Achtsamkeits-Mentor](https://docs.claude.com/de/resources/prompt-library/mindfulness-mentor.md): Führe den Benutzer durch Achtsamkeitsübungen und Techniken zur Stressreduzierung.
- [Stimmungs-Farbgeber](https://docs.claude.com/de/resources/prompt-library/mood-colorizer.md): Verwandelt Textbeschreibungen von Stimmungen in entsprechende HEX-Codes.
- [Motivierende Muse](https://docs.claude.com/de/resources/prompt-library/motivational-muse.md): Biete personalisierte motivierende Nachrichten und Affirmationen basierend auf Benutzereingaben.
- [Neologismus-Ersteller](https://docs.claude.com/de/resources/prompt-library/neologism-creator.md): Erfinde neue Wörter und liefere ihre Definitionen basierend auf benutzerdefinierten Konzepten oder Ideen.
- [Perspektiven-Grübler](https://docs.claude.com/de/resources/prompt-library/perspectives-ponderer.md): Wägen Sie die Vor- und Nachteile eines vom Benutzer bereitgestellten Themas ab.
- [Philosophische Betrachtungen](https://docs.claude.com/de/resources/prompt-library/philosophical-musings.md): Führen Sie tiefgreifende philosophische Diskussionen und Gedankenexperimente.
- [PII-Bereiniger](https://docs.claude.com/de/resources/prompt-library/pii-purifier.md): Automatische Erkennung und Entfernung von personenbezogenen Daten (PII) aus Textdokumenten.
- [Polyglotte Superkräfte](https://docs.claude.com/de/resources/prompt-library/polyglot-superpowers.md): Übersetze Text aus jeder Sprache in jede andere Sprache.
- [Portmanteau-Dichter](https://docs.claude.com/de/resources/prompt-library/portmanteau-poet.md): Verschmelze zwei Wörter miteinander, um ein neues, bedeutungsvolles Portmanteau zu schaffen.
- [Produktnamen-Profi](https://docs.claude.com/de/resources/prompt-library/product-naming-pro.md): Erstelle eingängige Produktnamen aus Beschreibungen und Schlüsselwörtern.
- [Prosa-Polierer](https://docs.claude.com/de/resources/prompt-library/prose-polisher.md): Verfeinern und verbessern Sie schriftliche Inhalte mit fortgeschrittenen Lektoratstechniken und Vorschlägen.
- [Wortspiel-Experte](https://docs.claude.com/de/resources/prompt-library/pun-dit.md): Generiere clevere Wortspiele und Wortwitz basierend auf jedem gegebenen Thema.
- [Python Bug-Beheber](https://docs.claude.com/de/resources/prompt-library/python-bug-buster.md): Erkennen und beheben Sie Bugs in Python-Code.
- [Bewertungsklassifizierer](https://docs.claude.com/de/resources/prompt-library/review-classifier.md): Kategorisiere Feedback in vorgegebene Tags und Kategorisierungen.
- [Rätsel mir das](https://docs.claude.com/de/resources/prompt-library/riddle-me-this.md): Generiere Rätsel und führe den Benutzer zu den Lösungen.
- [Science-Fiction-Szenario-Simulator](https://docs.claude.com/de/resources/prompt-library/sci-fi-scenario-simulator.md): Diskutieren Sie mit dem Benutzer verschiedene Science-Fiction-Szenarien und die damit verbundenen Herausforderungen und Überlegungen.
- [Vereinfacher für die zweite Klasse](https://docs.claude.com/de/resources/prompt-library/second-grade-simplifier.md): Mache komplexe Texte für junge Lernende leicht verständlich.
- [Gleichnis-Gelehrter](https://docs.claude.com/de/resources/prompt-library/simile-savant.md): Generiere Gleichnisse aus einfachen Beschreibungen.
- [Sokratischer Weise](https://docs.claude.com/de/resources/prompt-library/socratic-sage.md): Führe ein Gespräch im sokratischen Stil über ein vom Benutzer vorgegebenes Thema.
- [Tabellenkalkulations-Zauberer](https://docs.claude.com/de/resources/prompt-library/spreadsheet-sorcerer.md): Generiere CSV-Tabellenkalkulationen mit verschiedenen Datentypen.
- [SQL-Zauberer](https://docs.claude.com/de/resources/prompt-library/sql-sorcerer.md): Verwandeln Sie alltägliche Sprache in SQL-Abfragen.
- [Geschichtenerzähler-Begleiter](https://docs.claude.com/de/resources/prompt-library/storytelling-sidekick.md): Erstelle gemeinsam mit dem Benutzer fesselnde Geschichten und biete Wendungen in der Handlung und Charakterentwicklung.
- [Zeitreise-Berater](https://docs.claude.com/de/resources/prompt-library/time-travel-consultant.md): Hilf dem Benutzer bei der Navigation durch hypothetische Zeitreise-Szenarien und deren Auswirkungen.
- [Zungenbrecher](https://docs.claude.com/de/resources/prompt-library/tongue-twister.md): Erstelle herausfordernde Zungenbrecher.
- [Trivia-Generator](https://docs.claude.com/de/resources/prompt-library/trivia-generator.md): Generiere Trivia-Fragen zu einer Vielzahl von Themen und biete Hinweise bei Bedarf.
- [Tweet-Ton-Detektor](https://docs.claude.com/de/resources/prompt-library/tweet-tone-detector.md): Erkennen Sie den Ton und die Stimmung hinter Tweets.
- [VR-Fitness-Innovator](https://docs.claude.com/de/resources/prompt-library/vr-fitness-innovator.md): Brainstorming kreativer Ideen für Virtual-Reality-Fitnessspiele.
- [Website-Assistent](https://docs.claude.com/de/resources/prompt-library/website-wizard.md): Erstelle einseitige Websites basierend auf Benutzerspezifikationen.
- [Schnellstarts](https://docs.claude.com/de/resources/quickstarts.md): Schnell mit Claude und der Claude API loslegen
- [Systemstatus](https://docs.claude.com/de/resources/status.md)
- [Get API Key](https://docs.claude.com/en/api/admin-api/apikeys/get-api-key.md)
- [List API Keys](https://docs.claude.com/en/api/admin-api/apikeys/list-api-keys.md)
- [Update API Keys](https://docs.claude.com/en/api/admin-api/apikeys/update-api-key.md)
- [Get Claude Code Usage Report](https://docs.claude.com/en/api/admin-api/claude-code/get-claude-code-usage-report.md): Retrieve daily aggregated usage metrics for Claude Code users.
Enables organizations to analyze developer productivity and build custom dashboards.
- [Create Invite](https://docs.claude.com/en/api/admin-api/invites/create-invite.md)
- [Delete Invite](https://docs.claude.com/en/api/admin-api/invites/delete-invite.md)
- [Get Invite](https://docs.claude.com/en/api/admin-api/invites/get-invite.md)
- [List Invites](https://docs.claude.com/en/api/admin-api/invites/list-invites.md)
- [Get Organization Info](https://docs.claude.com/en/api/admin-api/organization/get-me.md)
- [Get Cost Report](https://docs.claude.com/en/api/admin-api/usage-cost/get-cost-report.md)
- [Get Usage Report for the Messages API](https://docs.claude.com/en/api/admin-api/usage-cost/get-messages-usage-report.md)
- [Get User](https://docs.claude.com/en/api/admin-api/users/get-user.md)
- [List Users](https://docs.claude.com/en/api/admin-api/users/list-users.md)
- [Remove User](https://docs.claude.com/en/api/admin-api/users/remove-user.md)
- [Update User](https://docs.claude.com/en/api/admin-api/users/update-user.md)
- [Add Workspace Member](https://docs.claude.com/en/api/admin-api/workspace_members/create-workspace-member.md)
- [Delete Workspace Member](https://docs.claude.com/en/api/admin-api/workspace_members/delete-workspace-member.md)
- [Get Workspace Member](https://docs.claude.com/en/api/admin-api/workspace_members/get-workspace-member.md)
- [List Workspace Members](https://docs.claude.com/en/api/admin-api/workspace_members/list-workspace-members.md)
- [Update Workspace Member](https://docs.claude.com/en/api/admin-api/workspace_members/update-workspace-member.md)
- [Archive Workspace](https://docs.claude.com/en/api/admin-api/workspaces/archive-workspace.md)
- [Create Workspace](https://docs.claude.com/en/api/admin-api/workspaces/create-workspace.md)
- [Get Workspace](https://docs.claude.com/en/api/admin-api/workspaces/get-workspace.md)
- [List Workspaces](https://docs.claude.com/en/api/admin-api/workspaces/list-workspaces.md)
- [Update Workspace](https://docs.claude.com/en/api/admin-api/workspaces/update-workspace.md)
- [Admin API overview](https://docs.claude.com/en/api/administration-api.md)
- [Beta headers](https://docs.claude.com/en/api/beta-headers.md): Documentation for using beta headers with the Claude API
- [Cancel a Message Batch](https://docs.claude.com/en/api/canceling-message-batches.md): Batches may be canceled any time before processing ends. Once cancellation is initiated, the batch enters a `canceling` state, at which time the system may complete any in-progress, non-interruptible requests before finalizing cancellation.
The number of canceled requests is specified in `request_counts`. To determine which requests were canceled, check the individual results within the batch. Note that cancellation may not result in any canceled requests if they were non-interruptible.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Claude Code Analytics API](https://docs.claude.com/en/api/claude-code-analytics-api.md): Programmatically access your organization's Claude Code usage analytics and productivity metrics with the Claude Code Analytics Admin API.
- [Amazon Bedrock API](https://docs.claude.com/en/api/claude-on-amazon-bedrock.md): Anthropic's Claude models are now generally available through Amazon Bedrock.
- [Vertex AI API](https://docs.claude.com/en/api/claude-on-vertex-ai.md): Anthropic's Claude models are now generally available through [Vertex AI](https://cloud.google.com/vertex-ai).
- [Client SDKs](https://docs.claude.com/en/api/client-sdks.md): We provide client libraries in a number of popular languages that make it easier to work with the Claude API.
- [Create a Message Batch](https://docs.claude.com/en/api/creating-message-batches.md): Send a batch of Message creation requests.
The Message Batches API can be used to process multiple Messages API requests at once. Once a Message Batch is created, it begins processing immediately. Batches can take up to 24 hours to complete.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Delete a Message Batch](https://docs.claude.com/en/api/deleting-message-batches.md): Delete a Message Batch.
Message Batches can only be deleted once they've finished processing. If you'd like to delete an in-progress batch, you must first cancel it.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Errors](https://docs.claude.com/en/api/errors.md)
- [Download a File](https://docs.claude.com/en/api/files-content.md): Download the contents of a Claude generated file
- [Create a File](https://docs.claude.com/en/api/files-create.md): Upload a file
- [Delete a File](https://docs.claude.com/en/api/files-delete.md): Make a file inaccessible through the API
- [List Files](https://docs.claude.com/en/api/files-list.md): List files within a workspace
- [Get File Metadata](https://docs.claude.com/en/api/files-metadata.md)
- [Getting help](https://docs.claude.com/en/api/getting-help.md): We've tried to provide the answers to the most common questions in these docs. However, if you need further technical support using Claude, the Claude API, or any of our products, you may reach our support team at [support.claude.com](https://support.claude.com).
- [Handling stop reasons](https://docs.claude.com/en/api/handling-stop-reasons.md)
- [IP addresses](https://docs.claude.com/en/api/ip-addresses.md): Anthropic services use fixed IP addresses for both inbound and outbound connections. You can use these addresses to configure your firewall rules for secure access to the Claude API and Console. These addresses will not change without notice.
- [List Message Batches](https://docs.claude.com/en/api/listing-message-batches.md): List all Message Batches within a Workspace. Most recently created batches are returned first.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Messages](https://docs.claude.com/en/api/messages.md): Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless multi-turn conversations.
Learn more about the Messages API in our [user guide](/en/docs/initial-setup)
- [Message Batches examples](https://docs.claude.com/en/api/messages-batch-examples.md): Example usage for the Message Batches API
- [Count Message tokens](https://docs.claude.com/en/api/messages-count-tokens.md): Count the number of tokens in a Message.
The Token Count API can be used to count the number of tokens in a Message, including tools, images, and documents, without creating it.
Learn more about token counting in our [user guide](/en/docs/build-with-claude/token-counting)
- [Messages examples](https://docs.claude.com/en/api/messages-examples.md): Request and response examples for the Messages API
- [Migrating from Text Completions](https://docs.claude.com/en/api/migrating-from-text-completions-to-messages.md): Migrating from Text Completions to Messages
- [Get a Model](https://docs.claude.com/en/api/models.md): Get a specific model.
The Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.
- [List Models](https://docs.claude.com/en/api/models-list.md): List available models.
The Models API response can be used to determine which models are available for use in the API. More recently released models are listed first.
- [OpenAI SDK compatibility](https://docs.claude.com/en/api/openai-sdk.md): Anthropic provides a compatibility layer that enables you to use the OpenAI SDK to test the Claude API. With a few code changes, you can quickly evaluate Anthropic model capabilities.
- [Overview](https://docs.claude.com/en/api/overview.md)
- [Generate a prompt](https://docs.claude.com/en/api/prompt-tools-generate.md): Generate a well-written prompt
- [Improve a prompt](https://docs.claude.com/en/api/prompt-tools-improve.md): Create a new-and-improved prompt guided by feedback
- [Templatize a prompt](https://docs.claude.com/en/api/prompt-tools-templatize.md): Templatize a prompt by indentifying and extracting variables
- [Rate limits](https://docs.claude.com/en/api/rate-limits.md): To mitigate misuse and manage capacity on our API, we have implemented limits on how much an organization can use the Claude API.
- [Retrieve Message Batch Results](https://docs.claude.com/en/api/retrieving-message-batch-results.md): Streams the results of a Message Batch as a `.jsonl` file.
Each line in the file is a JSON object containing the result of a single request in the Message Batch. Results are not guaranteed to be in the same order as requests. Use the `custom_id` field to match results to requests.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Retrieve a Message Batch](https://docs.claude.com/en/api/retrieving-message-batches.md): This endpoint is idempotent and can be used to poll for Message Batch completion. To access the results of a Message Batch, make a request to the `results_url` field in the response.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Service tiers](https://docs.claude.com/en/api/service-tiers.md): Different tiers of service allow you to balance availability, performance, and predictable costs based on your application's needs.
- [Supported regions](https://docs.claude.com/en/api/supported-regions.md): Here are the countries, regions, and territories we can currently support access from:
- [Usage and Cost API](https://docs.claude.com/en/api/usage-cost-api.md): Programmatically access your organization's API usage and cost data with the Usage & Cost Admin API.
- [Versions](https://docs.claude.com/en/api/versioning.md): When making API requests, you must send an `anthropic-version` request header. For example, `anthropic-version: 2023-06-01`. If you are using our [client SDKs](/en/api/client-sdks), this is handled for you automatically.
- [Glossary](https://docs.claude.com/en/docs/about-claude/glossary.md): These concepts are not unique to Anthropic’s language models, but we present a brief summary of key terms below.
- [Model deprecations](https://docs.claude.com/en/docs/about-claude/model-deprecations.md)
- [Choosing the right model](https://docs.claude.com/en/docs/about-claude/models/choosing-a-model.md): Selecting the optimal Claude model for your application involves balancing three key considerations: capabilities, speed, and cost. This guide helps you make an informed decision based on your specific requirements.
- [Migrating to Claude 4](https://docs.claude.com/en/docs/about-claude/models/migrating-to-claude-4.md)
- [Models overview](https://docs.claude.com/en/docs/about-claude/models/overview.md): Claude is a family of state-of-the-art large language models developed by Anthropic. This guide introduces our models and compares their performance with legacy models.
- [Pricing](https://docs.claude.com/en/docs/about-claude/pricing.md): Learn about Anthropic's pricing structure for models and features
- [Content moderation](https://docs.claude.com/en/docs/about-claude/use-case-guides/content-moderation.md): Content moderation is a critical aspect of maintaining a safe, respectful, and productive environment in digital applications. In this guide, we'll discuss how Claude can be used to moderate content within your digital application.
- [Customer support agent](https://docs.claude.com/en/docs/about-claude/use-case-guides/customer-support-chat.md): This guide walks through how to leverage Claude's advanced conversational capabilities to handle customer inquiries in real time, providing 24/7 support, reducing wait times, and managing high support volumes with accurate responses and positive interactions.
- [Legal summarization](https://docs.claude.com/en/docs/about-claude/use-case-guides/legal-summarization.md): This guide walks through how to leverage Claude's advanced natural language processing capabilities to efficiently summarize legal documents, extracting key information and expediting legal research. With Claude, you can streamline the review of contracts, litigation prep, and regulatory work, saving time and ensuring accuracy in your legal processes.
- [Guides to common use cases](https://docs.claude.com/en/docs/about-claude/use-case-guides/overview.md)
- [Ticket routing](https://docs.claude.com/en/docs/about-claude/use-case-guides/ticket-routing.md): This guide walks through how to harness Claude's advanced natural language understanding capabilities to classify customer support tickets at scale based on customer intent, urgency, prioritization, customer profile, and more.
- [Google Sheets add-on](https://docs.claude.com/en/docs/agents-and-tools/claude-for-sheets.md): The [Claude for Sheets extension](https://workspace.google.com/marketplace/app/claude%5Ffor%5Fsheets/909417792257) integrates Claude into Google Sheets, allowing you to execute interactions with Claude directly in cells.
- [MCP connector](https://docs.claude.com/en/docs/agents-and-tools/mcp-connector.md)
- [Remote MCP servers](https://docs.claude.com/en/docs/agents-and-tools/remote-mcp-servers.md)
- [Bash tool](https://docs.claude.com/en/docs/agents-and-tools/tool-use/bash-tool.md)
- [Code execution tool](https://docs.claude.com/en/docs/agents-and-tools/tool-use/code-execution-tool.md)
- [Computer use tool](https://docs.claude.com/en/docs/agents-and-tools/tool-use/computer-use-tool.md)
- [Fine-grained tool streaming](https://docs.claude.com/en/docs/agents-and-tools/tool-use/fine-grained-tool-streaming.md)
- [How to implement tool use](https://docs.claude.com/en/docs/agents-and-tools/tool-use/implement-tool-use.md)
- [Tool use with Claude](https://docs.claude.com/en/docs/agents-and-tools/tool-use/overview.md)
- [Text editor tool](https://docs.claude.com/en/docs/agents-and-tools/tool-use/text-editor-tool.md)
- [Token-efficient tool use](https://docs.claude.com/en/docs/agents-and-tools/tool-use/token-efficient-tool-use.md)
- [Web fetch tool](https://docs.claude.com/en/docs/agents-and-tools/tool-use/web-fetch-tool.md)
- [Web search tool](https://docs.claude.com/en/docs/agents-and-tools/tool-use/web-search-tool.md)
- [Batch processing](https://docs.claude.com/en/docs/build-with-claude/batch-processing.md)
- [Citations](https://docs.claude.com/en/docs/build-with-claude/citations.md)
- [Context windows](https://docs.claude.com/en/docs/build-with-claude/context-windows.md)
- [Embeddings](https://docs.claude.com/en/docs/build-with-claude/embeddings.md): Text embeddings are numerical representations of text that enable measuring semantic similarity. This guide introduces embeddings, their applications, and how to use embedding models for tasks like search, recommendations, and anomaly detection.
- [Building with extended thinking](https://docs.claude.com/en/docs/build-with-claude/extended-thinking.md)
- [Files API](https://docs.claude.com/en/docs/build-with-claude/files.md)
- [Multilingual support](https://docs.claude.com/en/docs/build-with-claude/multilingual-support.md): Claude excels at tasks across multiple languages, maintaining strong cross-lingual performance relative to English.
- [Features overview](https://docs.claude.com/en/docs/build-with-claude/overview.md): Explore Claude's advanced features and capabilities.
- [PDF support](https://docs.claude.com/en/docs/build-with-claude/pdf-support.md): Process PDFs with Claude. Extract text, analyze charts, and understand visual content from your documents.
- [Prompt caching](https://docs.claude.com/en/docs/build-with-claude/prompt-caching.md)
- [Be clear, direct, and detailed](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/be-clear-and-direct.md)
- [Let Claude think (chain of thought prompting) to increase performance](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/chain-of-thought.md)
- [Chain complex prompts for stronger performance](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/chain-prompts.md)
- [Claude 4 prompt engineering best practices](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/claude-4-best-practices.md)
- [Extended thinking tips](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/extended-thinking-tips.md)
- [Long context prompting tips](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/long-context-tips.md)
- [Use examples (multishot prompting) to guide Claude's behavior](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/multishot-prompting.md)
- [Prompt engineering overview](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/overview.md)
- [Prefill Claude's response for greater output control](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/prefill-claudes-response.md)
- [Automatically generate first draft prompt templates](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/prompt-generator.md)
- [Use our prompt improver to optimize your prompts](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/prompt-improver.md)
- [Use prompt templates and variables](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/prompt-templates-and-variables.md)
- [Giving Claude a role with a system prompt](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/system-prompts.md)
- [Use XML tags to structure your prompts](https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/use-xml-tags.md)
- [Search results](https://docs.claude.com/en/docs/build-with-claude/search-results.md): Enable natural citations for RAG applications by providing search results with source attribution
- [Streaming Messages](https://docs.claude.com/en/docs/build-with-claude/streaming.md)
- [Token counting](https://docs.claude.com/en/docs/build-with-claude/token-counting.md)
- [Vision](https://docs.claude.com/en/docs/build-with-claude/vision.md): The Claude 3 and 4 families of models comes with new vision capabilities that allow Claude to understand and analyze images, opening up exciting possibilities for multimodal interaction.
- [Claude Code on Amazon Bedrock](https://docs.claude.com/en/docs/claude-code/amazon-bedrock.md): Learn about configuring Claude Code through Amazon Bedrock, including setup, IAM configuration, and troubleshooting.
- [Analytics](https://docs.claude.com/en/docs/claude-code/analytics.md): View detailed usage insights and productivity metrics for your organization's Claude Code deployment.
- [CLI reference](https://docs.claude.com/en/docs/claude-code/cli-reference.md): Complete reference for Claude Code command-line interface, including commands and flags.
- [Common workflows](https://docs.claude.com/en/docs/claude-code/common-workflows.md): Learn about common workflows with Claude Code.
- [Manage costs effectively](https://docs.claude.com/en/docs/claude-code/costs.md): Learn how to track and optimize token usage and costs when using Claude Code.
- [Data usage](https://docs.claude.com/en/docs/claude-code/data-usage.md): Learn about Anthropic's data usage policies for Claude
- [Development containers](https://docs.claude.com/en/docs/claude-code/devcontainer.md): Learn about the Claude Code development container for teams that need consistent, secure environments.
- [Claude Code GitHub Actions](https://docs.claude.com/en/docs/claude-code/github-actions.md): Learn about integrating Claude Code into your development workflow with Claude Code GitHub Actions
- [Claude Code GitLab CI/CD](https://docs.claude.com/en/docs/claude-code/gitlab-ci-cd.md): Learn about integrating Claude Code into your development workflow with GitLab CI/CD
- [Claude Code on Google Vertex AI](https://docs.claude.com/en/docs/claude-code/google-vertex-ai.md): Learn about configuring Claude Code through Google Vertex AI, including setup, IAM configuration, and troubleshooting.
- [Hooks reference](https://docs.claude.com/en/docs/claude-code/hooks.md): This page provides reference documentation for implementing hooks in Claude Code.
- [Get started with Claude Code hooks](https://docs.claude.com/en/docs/claude-code/hooks-guide.md): Learn how to customize and extend Claude Code's behavior by registering shell commands
- [Identity and Access Management](https://docs.claude.com/en/docs/claude-code/iam.md): Learn how to configure user authentication, authorization, and access controls for Claude Code in your organization.
- [Add Claude Code to your IDE](https://docs.claude.com/en/docs/claude-code/ide-integrations.md): Learn how to add Claude Code to your favorite IDE
- [Interactive mode](https://docs.claude.com/en/docs/claude-code/interactive-mode.md): Complete reference for keyboard shortcuts, input modes, and interactive features in Claude Code sessions.
- [Legal and compliance](https://docs.claude.com/en/docs/claude-code/legal-and-compliance.md): Legal agreements, compliance certifications, and security information for Claude Code.
- [LLM gateway configuration](https://docs.claude.com/en/docs/claude-code/llm-gateway.md): Learn how to configure Claude Code with LLM gateway solutions, including LiteLLM setup, authentication methods, and enterprise features like usage tracking and budget management.
- [Connect Claude Code to tools via MCP](https://docs.claude.com/en/docs/claude-code/mcp.md): Learn how to connect Claude Code to your tools with the Model Context Protocol.
- [Manage Claude's memory](https://docs.claude.com/en/docs/claude-code/memory.md): Learn how to manage Claude Code's memory across sessions with different memory locations and best practices.
- [Model configuration](https://docs.claude.com/en/docs/claude-code/model-config.md): Learn about the Claude Code model configuration, including model aliases like `opusplan`
- [Monitoring](https://docs.claude.com/en/docs/claude-code/monitoring-usage.md): Learn how to enable and configure OpenTelemetry for Claude Code.
- [Enterprise network configuration](https://docs.claude.com/en/docs/claude-code/network-config.md): Configure Claude Code for enterprise environments with proxy servers, custom Certificate Authorities (CA), and mutual Transport Layer Security (mTLS) authentication.
- [Output styles](https://docs.claude.com/en/docs/claude-code/output-styles.md): Adapt Claude Code for uses beyond software engineering
- [Claude Code overview](https://docs.claude.com/en/docs/claude-code/overview.md): Learn about Claude Code, Anthropic's agentic coding tool that lives in your terminal and helps you turn ideas into code faster than ever before.
- [Quickstart](https://docs.claude.com/en/docs/claude-code/quickstart.md): Welcome to Claude Code!
- [Custom Tools](https://docs.claude.com/en/docs/claude-code/sdk/custom-tools.md): Build and integrate custom tools to extend Claude Code SDK functionality
- [Modifying system prompts](https://docs.claude.com/en/docs/claude-code/sdk/modifying-system-prompts.md): Learn how to customize Claude's behavior by modifying system prompts using three approaches - output styles, appendSystemPrompt, and customSystemPrompt.
- [Tracking Costs and Usage](https://docs.claude.com/en/docs/claude-code/sdk/sdk-cost-tracking.md): Understand and track token usage for billing in the Claude Code SDK
- [Headless mode](https://docs.claude.com/en/docs/claude-code/sdk/sdk-headless.md): Run Claude Code programmatically without interactive UI
- [MCP in the SDK](https://docs.claude.com/en/docs/claude-code/sdk/sdk-mcp.md): Extend Claude Code with custom tools using Model Context Protocol servers
- [Overview](https://docs.claude.com/en/docs/claude-code/sdk/sdk-overview.md): Build custom AI agents with the Claude Code SDK
- [Handling Permissions](https://docs.claude.com/en/docs/claude-code/sdk/sdk-permissions.md): Control tool usage and permissions in the Claude Code SDK
- [Python SDK reference](https://docs.claude.com/en/docs/claude-code/sdk/sdk-python.md): Complete API reference for the Claude Code Python SDK, including all functions, types, and classes.
- [Session Management](https://docs.claude.com/en/docs/claude-code/sdk/sdk-sessions.md): Understanding how the Claude Code SDK handles sessions and session resumption
- [Slash Commands in the SDK](https://docs.claude.com/en/docs/claude-code/sdk/sdk-slash-commands.md): Learn how to use slash commands to control Claude Code sessions through the SDK
- [TypeScript SDK reference](https://docs.claude.com/en/docs/claude-code/sdk/sdk-typescript.md): Complete API reference for the Claude Code TypeScript SDK, including all functions, types, and interfaces.
- [Streaming Input](https://docs.claude.com/en/docs/claude-code/sdk/streaming-vs-single-mode.md): Understanding the two input modes for Claude Code SDK and when to use each
- [Subagents in the SDK](https://docs.claude.com/en/docs/claude-code/sdk/subagents.md): Working with subagents in the Claude Code SDK
- [Todo Lists](https://docs.claude.com/en/docs/claude-code/sdk/todo-tracking.md): Track and display todos using the Claude Code SDK for organized task management
- [Security](https://docs.claude.com/en/docs/claude-code/security.md): Learn about Claude Code's security safeguards and best practices for safe usage.
- [Claude Code settings](https://docs.claude.com/en/docs/claude-code/settings.md): Configure Claude Code with global and project-level settings, and environment variables.
- [Set up Claude Code](https://docs.claude.com/en/docs/claude-code/setup.md): Install, authenticate, and start using Claude Code on your development machine.
- [Slash commands](https://docs.claude.com/en/docs/claude-code/slash-commands.md): Control Claude's behavior during an interactive session with slash commands.
- [Status line configuration](https://docs.claude.com/en/docs/claude-code/statusline.md): Create a custom status line for Claude Code to display contextual information
- [Subagents](https://docs.claude.com/en/docs/claude-code/sub-agents.md): Create and use specialized AI subagents in Claude Code for task-specific workflows and improved context management.
- [Optimize your terminal setup](https://docs.claude.com/en/docs/claude-code/terminal-config.md): Claude Code works best when your terminal is properly configured. Follow these guidelines to optimize your experience.
- [Enterprise deployment overview](https://docs.claude.com/en/docs/claude-code/third-party-integrations.md): Learn how Claude Code can integrate with various third-party services and infrastructure to meet enterprise deployment requirements.
- [Troubleshooting](https://docs.claude.com/en/docs/claude-code/troubleshooting.md): Discover solutions to common issues with Claude Code installation and usage.
- [Get started with Claude](https://docs.claude.com/en/docs/get-started.md): Make your first API call to Claude and build a simple web search assistant
- [Intro to Claude](https://docs.claude.com/en/docs/intro.md): Claude is a highly performant, trustworthy, and intelligent AI platform built by Anthropic. Claude excels at tasks involving language, reasoning, analysis, coding, and more.
- [Anthropic Privacy Policy](https://docs.claude.com/en/docs/legal-center/privacy.md)
- [Security and compliance](https://docs.claude.com/en/docs/legal-center/security-compliance.md)
- [Model Context Protocol (MCP)](https://docs.claude.com/en/docs/mcp.md)
- [Building with Claude](https://docs.claude.com/en/docs/overview.md): Claude is a family of [highly performant and intelligent AI models](/en/docs/about-claude/models) built by Anthropic. While Claude is powerful and extensible, it's also the most trustworthy and reliable AI available. It follows critical protocols, makes fewer mistakes, and is resistant to jailbreaks—allowing [enterprise customers](https://claude.com/customers) to build the safest AI-powered applications at scale.
- [Define your success criteria](https://docs.claude.com/en/docs/test-and-evaluate/define-success.md)
- [Create strong empirical evaluations](https://docs.claude.com/en/docs/test-and-evaluate/develop-tests.md)
- [Using the Evaluation Tool](https://docs.claude.com/en/docs/test-and-evaluate/eval-tool.md): The [Claude Console](https://console.anthropic.com/dashboard) features an **Evaluation tool** that allows you to test your prompts under various scenarios.
- [Streaming refusals](https://docs.claude.com/en/docs/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals.md)
- [Increase output consistency (JSON mode)](https://docs.claude.com/en/docs/test-and-evaluate/strengthen-guardrails/increase-consistency.md)
- [Keep Claude in character with role prompting and prefilling](https://docs.claude.com/en/docs/test-and-evaluate/strengthen-guardrails/keep-claude-in-character.md)
- [Mitigate jailbreaks and prompt injections](https://docs.claude.com/en/docs/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks.md)
- [Reduce hallucinations](https://docs.claude.com/en/docs/test-and-evaluate/strengthen-guardrails/reduce-hallucinations.md)
- [Reducing latency](https://docs.claude.com/en/docs/test-and-evaluate/strengthen-guardrails/reduce-latency.md)
- [Reduce prompt leak](https://docs.claude.com/en/docs/test-and-evaluate/strengthen-guardrails/reduce-prompt-leak.md)
- [null](https://docs.claude.com/en/home.md)
- [Claude Developer Platform](https://docs.claude.com/en/release-notes/api.md): Follow along with updates to the Claude API and the Claude Console.
- [Claude Apps](https://docs.claude.com/en/release-notes/claude-apps.md): Follow along with updates across Anthropic's Claude applications.
- [Claude Code](https://docs.claude.com/en/release-notes/claude-code.md): Stay up to date with the latest features and improvements to Claude Code
- [Overview](https://docs.claude.com/en/release-notes/overview.md): Follow along with updates across Anthropic's products and services.
- [System Prompts](https://docs.claude.com/en/release-notes/system-prompts.md): See updates to the core system prompts on [Claude.ai](https://www.claude.ai) and the Claude [iOS](http://anthropic.com/ios) and [Android](http://anthropic.com/android) apps.
- [Claude 3.7 system card](https://docs.claude.com/en/resources/claude-3-7-system-card.md)
- [Claude 3 model card](https://docs.claude.com/en/resources/claude-3-model-card.md)
- [Claude 4 System Card](https://docs.claude.com/en/resources/claude-4-system-card.md)
- [Anthropic Courses](https://docs.claude.com/en/resources/courses.md)
- [null](https://docs.claude.com/en/resources/overview.md)
- [Adaptive editor](https://docs.claude.com/en/resources/prompt-library/adaptive-editor.md): Rewrite text following user-given instructions, such as with a different tone, audience, or style.
- [Airport code analyst](https://docs.claude.com/en/resources/prompt-library/airport-code-analyst.md): Find and extract airport codes from text.
- [Alien anthropologist](https://docs.claude.com/en/resources/prompt-library/alien-anthropologist.md): Analyze human culture and customs from the perspective of an alien anthropologist.
- [Alliteration alchemist](https://docs.claude.com/en/resources/prompt-library/alliteration-alchemist.md): Generate alliterative phrases and sentences for any given subject.
- [Babel's broadcasts](https://docs.claude.com/en/resources/prompt-library/babels-broadcasts.md): Create compelling product announcement tweets in the world's 10 most spoken languages.
- [Brand builder](https://docs.claude.com/en/resources/prompt-library/brand-builder.md): Craft a design brief for a holistic brand identity.
- [Career coach](https://docs.claude.com/en/resources/prompt-library/career-coach.md): Engage in role-play conversations with an AI career coach.
- [Cite your sources](https://docs.claude.com/en/resources/prompt-library/cite-your-sources.md): Get answers to questions about a document's content with relevant citations supporting the response.
- [Code clarifier](https://docs.claude.com/en/resources/prompt-library/code-clarifier.md): Simplify and explain complex code in plain language.
- [Code consultant](https://docs.claude.com/en/resources/prompt-library/code-consultant.md): Suggest improvements to optimize Python code performance.
- [Corporate clairvoyant](https://docs.claude.com/en/resources/prompt-library/corporate-clairvoyant.md): Extract insights, identify risks, and distill key information from long corporate reports into a single memo.
- [Cosmic Keystrokes](https://docs.claude.com/en/resources/prompt-library/cosmic-keystrokes.md): Generate an interactive speed typing game in a single HTML file, featuring side-scrolling gameplay and Tailwind CSS styling.
- [CSV converter](https://docs.claude.com/en/resources/prompt-library/csv-converter.md): Convert data from various formats (JSON, XML, etc.) into properly formatted CSV files.
- [Culinary creator](https://docs.claude.com/en/resources/prompt-library/culinary-creator.md): Suggest recipe ideas based on the user's available ingredients and dietary preferences.
- [Data organizer](https://docs.claude.com/en/resources/prompt-library/data-organizer.md): Turn unstructured text into bespoke JSON tables.
- [Direction decoder](https://docs.claude.com/en/resources/prompt-library/direction-decoder.md): Transform natural language into step-by-step directions.
- [Dream interpreter](https://docs.claude.com/en/resources/prompt-library/dream-interpreter.md): Offer interpretations and insights into the symbolism of the user's dreams.
- [Efficiency estimator](https://docs.claude.com/en/resources/prompt-library/efficiency-estimator.md): Calculate the time complexity of functions and algorithms.
- [Email extractor](https://docs.claude.com/en/resources/prompt-library/email-extractor.md): Extract email addresses from a document into a JSON-formatted list.
- [Emoji encoder](https://docs.claude.com/en/resources/prompt-library/emoji-encoder.md): Convert plain text into fun and expressive emoji messages.
- [Ethical dilemma navigator](https://docs.claude.com/en/resources/prompt-library/ethical-dilemma-navigator.md): Help the user think through complex ethical dilemmas and provide different perspectives.
- [Excel formula expert](https://docs.claude.com/en/resources/prompt-library/excel-formula-expert.md): Create Excel formulas based on user-described calculations or data manipulations.
- [Function fabricator](https://docs.claude.com/en/resources/prompt-library/function-fabricator.md): Create Python functions based on detailed specifications.
- [Futuristic fashion advisor](https://docs.claude.com/en/resources/prompt-library/futuristic-fashion-advisor.md): Suggest avant-garde fashion trends and styles for the user's specific preferences.
- [Git gud](https://docs.claude.com/en/resources/prompt-library/git-gud.md): Generate appropriate Git commands based on user-described version control actions.
- [Google apps scripter](https://docs.claude.com/en/resources/prompt-library/google-apps-scripter.md): Generate Google Apps scripts to complete tasks based on user requirements.
- [Grading guru](https://docs.claude.com/en/resources/prompt-library/grading-guru.md): Compare and evaluate the quality of written texts based on user-defined criteria and standards.
- [Grammar genie](https://docs.claude.com/en/resources/prompt-library/grammar-genie.md): Transform grammatically incorrect sentences into proper English.
- [Hal the humorous helper](https://docs.claude.com/en/resources/prompt-library/hal-the-humorous-helper.md): Chat with a knowledgeable AI that has a sarcastic side.
- [Idiom illuminator](https://docs.claude.com/en/resources/prompt-library/idiom-illuminator.md): Explain the meaning and origin of common idioms and proverbs.
- [Interview question crafter](https://docs.claude.com/en/resources/prompt-library/interview-question-crafter.md): Generate questions for interviews.
- [LaTeX legend](https://docs.claude.com/en/resources/prompt-library/latex-legend.md): Write LaTeX documents, generating code for mathematical equations, tables, and more.
- [Lesson planner](https://docs.claude.com/en/resources/prompt-library/lesson-planner.md): Craft in depth lesson plans on any subject.
- [Prompt Library](https://docs.claude.com/en/resources/prompt-library/library.md)
- [Master moderator](https://docs.claude.com/en/resources/prompt-library/master-moderator.md): Evaluate user inputs for potential harmful or illegal content.
- [Meeting scribe](https://docs.claude.com/en/resources/prompt-library/meeting-scribe.md): Distill meetings into concise summaries including discussion topics, key takeaways, and action items.
- [Memo maestro](https://docs.claude.com/en/resources/prompt-library/memo-maestro.md): Compose comprehensive company memos based on key points.
- [Mindfulness mentor](https://docs.claude.com/en/resources/prompt-library/mindfulness-mentor.md): Guide the user through mindfulness exercises and techniques for stress reduction.
- [Mood colorizer](https://docs.claude.com/en/resources/prompt-library/mood-colorizer.md): Transform text descriptions of moods into corresponding HEX codes.
- [Motivational muse](https://docs.claude.com/en/resources/prompt-library/motivational-muse.md): Provide personalized motivational messages and affirmations based on user input.
- [Neologism creator](https://docs.claude.com/en/resources/prompt-library/neologism-creator.md): Invent new words and provide their definitions based on user-provided concepts or ideas.
- [Perspectives ponderer](https://docs.claude.com/en/resources/prompt-library/perspectives-ponderer.md): Weigh the pros and cons of a user-provided topic.
- [Philosophical musings](https://docs.claude.com/en/resources/prompt-library/philosophical-musings.md): Engage in deep philosophical discussions and thought experiments.
- [PII purifier](https://docs.claude.com/en/resources/prompt-library/pii-purifier.md): Automatically detect and remove personally identifiable information (PII) from text documents.
- [Polyglot superpowers](https://docs.claude.com/en/resources/prompt-library/polyglot-superpowers.md): Translate text from any language into any language.
- [Portmanteau poet](https://docs.claude.com/en/resources/prompt-library/portmanteau-poet.md): Blend two words together to create a new, meaningful portmanteau.
- [Product naming pro](https://docs.claude.com/en/resources/prompt-library/product-naming-pro.md): Create catchy product names from descriptions and keywords.
- [Prose polisher](https://docs.claude.com/en/resources/prompt-library/prose-polisher.md): Refine and improve written content with advanced copyediting techniques and suggestions.
- [Pun-dit](https://docs.claude.com/en/resources/prompt-library/pun-dit.md): Generate clever puns and wordplay based on any given topic.
- [Python bug buster](https://docs.claude.com/en/resources/prompt-library/python-bug-buster.md): Detect and fix bugs in Python code.
- [Review classifier](https://docs.claude.com/en/resources/prompt-library/review-classifier.md): Categorize feedback into pre-specified tags and categorizations.
- [Riddle me this](https://docs.claude.com/en/resources/prompt-library/riddle-me-this.md): Generate riddles and guide the user to the solutions.
- [Sci-fi scenario simulator](https://docs.claude.com/en/resources/prompt-library/sci-fi-scenario-simulator.md): Discuss with the user various science fiction scenarios and associated challenges and considerations.
- [Second-grade simplifier](https://docs.claude.com/en/resources/prompt-library/second-grade-simplifier.md): Make complex text easy for young learners to understand.
- [Simile savant](https://docs.claude.com/en/resources/prompt-library/simile-savant.md): Generate similes from basic descriptions.
- [Socratic sage](https://docs.claude.com/en/resources/prompt-library/socratic-sage.md): Engage in Socratic style conversation over a user-given topic.
- [Spreadsheet sorcerer](https://docs.claude.com/en/resources/prompt-library/spreadsheet-sorcerer.md): Generate CSV spreadsheets with various types of data.
- [SQL sorcerer](https://docs.claude.com/en/resources/prompt-library/sql-sorcerer.md): Transform everyday language into SQL queries.
- [Storytelling sidekick](https://docs.claude.com/en/resources/prompt-library/storytelling-sidekick.md): Collaboratively create engaging stories with the user, offering plot twists and character development.
- [Time travel consultant](https://docs.claude.com/en/resources/prompt-library/time-travel-consultant.md): Help the user navigate hypothetical time travel scenarios and their implications.
- [Tongue twister](https://docs.claude.com/en/resources/prompt-library/tongue-twister.md): Create challenging tongue twisters.
- [Trivia generator](https://docs.claude.com/en/resources/prompt-library/trivia-generator.md): Generate trivia questions on a wide range of topics and provide hints when needed.
- [Tweet tone detector](https://docs.claude.com/en/resources/prompt-library/tweet-tone-detector.md): Detect the tone and sentiment behind tweets.
- [VR fitness innovator](https://docs.claude.com/en/resources/prompt-library/vr-fitness-innovator.md): Brainstorm creative ideas for virtual reality fitness games.
- [Website wizard](https://docs.claude.com/en/resources/prompt-library/website-wizard.md): Create one-page websites based on user specifications.
- [Quickstarts](https://docs.claude.com/en/resources/quickstarts.md): Get started quickly with Claude and the Claude API
- [System status](https://docs.claude.com/en/resources/status.md)
- [Obtener Clave API](https://docs.claude.com/es/api/admin-api/apikeys/get-api-key.md)
- [Listar Claves API](https://docs.claude.com/es/api/admin-api/apikeys/list-api-keys.md)
- [Actualizar Claves API](https://docs.claude.com/es/api/admin-api/apikeys/update-api-key.md)
- [Obtener Reporte de Uso de Claude Code](https://docs.claude.com/es/api/admin-api/claude-code/get-claude-code-usage-report.md): Retrieve daily aggregated usage metrics for Claude Code users.
Enables organizations to analyze developer productivity and build custom dashboards.
- [Crear Invitación](https://docs.claude.com/es/api/admin-api/invites/create-invite.md)
- [Eliminar Invitación](https://docs.claude.com/es/api/admin-api/invites/delete-invite.md)
- [Obtener Invitación](https://docs.claude.com/es/api/admin-api/invites/get-invite.md)
- [Listar Invitaciones](https://docs.claude.com/es/api/admin-api/invites/list-invites.md)
- [Obtener Información de la Organización](https://docs.claude.com/es/api/admin-api/organization/get-me.md)
- [Obtener Informe de Costos](https://docs.claude.com/es/api/admin-api/usage-cost/get-cost-report.md)
- [Obtener Informe de Uso para la API de Mensajes](https://docs.claude.com/es/api/admin-api/usage-cost/get-messages-usage-report.md)
- [Obtener Usuario](https://docs.claude.com/es/api/admin-api/users/get-user.md)
- [Listar Usuarios](https://docs.claude.com/es/api/admin-api/users/list-users.md)
- [Eliminar Usuario](https://docs.claude.com/es/api/admin-api/users/remove-user.md)
- [Actualizar Usuario](https://docs.claude.com/es/api/admin-api/users/update-user.md)
- [Agregar Miembro del Espacio de Trabajo](https://docs.claude.com/es/api/admin-api/workspace_members/create-workspace-member.md)
- [Eliminar Miembro del Espacio de Trabajo](https://docs.claude.com/es/api/admin-api/workspace_members/delete-workspace-member.md)
- [Obtener Miembro del Espacio de Trabajo](https://docs.claude.com/es/api/admin-api/workspace_members/get-workspace-member.md)
- [Listar Miembros del Espacio de Trabajo](https://docs.claude.com/es/api/admin-api/workspace_members/list-workspace-members.md)
- [Actualizar Miembro del Espacio de Trabajo](https://docs.claude.com/es/api/admin-api/workspace_members/update-workspace-member.md)
- [Archivar Espacio de Trabajo](https://docs.claude.com/es/api/admin-api/workspaces/archive-workspace.md)
- [Crear Espacio de Trabajo](https://docs.claude.com/es/api/admin-api/workspaces/create-workspace.md)
- [Obtener Espacio de Trabajo](https://docs.claude.com/es/api/admin-api/workspaces/get-workspace.md)
- [Listar Espacios de Trabajo](https://docs.claude.com/es/api/admin-api/workspaces/list-workspaces.md)
- [Actualizar Espacio de Trabajo](https://docs.claude.com/es/api/admin-api/workspaces/update-workspace.md)
- [Descripción general de la API de administración](https://docs.claude.com/es/api/administration-api.md)
- [Encabezados beta](https://docs.claude.com/es/api/beta-headers.md): Documentación para usar encabezados beta con la API de Claude
- [Cancelar un Lote de Mensajes](https://docs.claude.com/es/api/canceling-message-batches.md): Batches may be canceled any time before processing ends. Once cancellation is initiated, the batch enters a `canceling` state, at which time the system may complete any in-progress, non-interruptible requests before finalizing cancellation.
The number of canceled requests is specified in `request_counts`. To determine which requests were canceled, check the individual results within the batch. Note that cancellation may not result in any canceled requests if they were non-interruptible.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [API de Análisis de Código Claude](https://docs.claude.com/es/api/claude-code-analytics-api.md): Accede programáticamente a los análisis de uso de Claude Code de tu organización y métricas de productividad con la API de Administración de Análisis de Código Claude.
- [API de Amazon Bedrock](https://docs.claude.com/es/api/claude-on-amazon-bedrock.md): Los modelos Claude de Anthropic ahora están disponibles de forma general a través de Amazon Bedrock.
- [API de Vertex AI](https://docs.claude.com/es/api/claude-on-vertex-ai.md): Los modelos Claude de Anthropic ahora están disponibles de forma general a través de [Vertex AI](https://cloud.google.com/vertex-ai).
- [SDKs de Cliente](https://docs.claude.com/es/api/client-sdks.md): Proporcionamos bibliotecas de cliente en varios lenguajes populares que facilitan el trabajo con la API de Claude.
- [Crear un Lote de Mensajes](https://docs.claude.com/es/api/creating-message-batches.md): Send a batch of Message creation requests.
The Message Batches API can be used to process multiple Messages API requests at once. Once a Message Batch is created, it begins processing immediately. Batches can take up to 24 hours to complete.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Eliminar un Lote de Mensajes](https://docs.claude.com/es/api/deleting-message-batches.md): Delete a Message Batch.
Message Batches can only be deleted once they've finished processing. If you'd like to delete an in-progress batch, you must first cancel it.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Errores](https://docs.claude.com/es/api/errors.md)
- [Descargar un Archivo](https://docs.claude.com/es/api/files-content.md): Descargar el contenido de un archivo generado por Claude
- [Crear un Archivo](https://docs.claude.com/es/api/files-create.md): Subir un archivo
- [Eliminar un Archivo](https://docs.claude.com/es/api/files-delete.md): Hacer que un archivo sea inaccesible a través de la API
- [Listar Archivos](https://docs.claude.com/es/api/files-list.md): Listar archivos dentro de un espacio de trabajo
- [Obtener Metadatos del Archivo](https://docs.claude.com/es/api/files-metadata.md)
- [Obtener ayuda](https://docs.claude.com/es/api/getting-help.md): Hemos tratado de proporcionar las respuestas a las preguntas más comunes en esta documentación. Sin embargo, si necesitas soporte técnico adicional usando Claude, la API de Claude, o cualquiera de nuestros productos, puedes contactar a nuestro equipo de soporte en [support.claude.com](https://support.claude.com).
- [Manejo de razones de parada](https://docs.claude.com/es/api/handling-stop-reasons.md): Aprende cómo manejar las diferentes razones de parada en las respuestas de la API de Messages de Claude.
- [Direcciones IP](https://docs.claude.com/es/api/ip-addresses.md): Los servicios de Anthropic utilizan direcciones IP fijas tanto para conexiones entrantes como salientes. Puedes usar estas direcciones para configurar las reglas de tu firewall para acceso seguro a la API de Claude y la Consola. Estas direcciones no cambiarán sin previo aviso.
- [Listar Lotes de Mensajes](https://docs.claude.com/es/api/listing-message-batches.md): List all Message Batches within a Workspace. Most recently created batches are returned first.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Mensajes](https://docs.claude.com/es/api/messages.md): Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless multi-turn conversations.
Learn more about the Messages API in our [user guide](/en/docs/initial-setup)
- [Ejemplos de Lotes de Mensajes](https://docs.claude.com/es/api/messages-batch-examples.md): Ejemplo de uso para la API de Lotes de Mensajes
- [Contar tokens de mensaje](https://docs.claude.com/es/api/messages-count-tokens.md): Count the number of tokens in a Message.
The Token Count API can be used to count the number of tokens in a Message, including tools, images, and documents, without creating it.
Learn more about token counting in our [user guide](/en/docs/build-with-claude/token-counting)
- [Ejemplos de mensajes](https://docs.claude.com/es/api/messages-examples.md): Ejemplos de solicitud y respuesta para la API de Mensajes
- [Migración desde Text Completions](https://docs.claude.com/es/api/migrating-from-text-completions-to-messages.md): Migración desde Text Completions a Messages
- [Obtener un Modelo](https://docs.claude.com/es/api/models.md): Get a specific model.
The Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.
- [Listar Modelos](https://docs.claude.com/es/api/models-list.md): List available models.
The Models API response can be used to determine which models are available for use in the API. More recently released models are listed first.
- [Compatibilidad con el SDK de OpenAI](https://docs.claude.com/es/api/openai-sdk.md): Anthropic proporciona una capa de compatibilidad que te permite usar el SDK de OpenAI para probar la API de Claude. Con algunos cambios de código, puedes evaluar rápidamente las capacidades del modelo de Anthropic.
- [Descripción general](https://docs.claude.com/es/api/overview.md): Descripción general de la API de Claude
- [Generar un prompt](https://docs.claude.com/es/api/prompt-tools-generate.md): Generar un prompt bien escrito
- [Mejorar un prompt](https://docs.claude.com/es/api/prompt-tools-improve.md): Crear un prompt nuevo y mejorado guiado por retroalimentación
- [Convertir un prompt en plantilla](https://docs.claude.com/es/api/prompt-tools-templatize.md): Convertir un prompt en plantilla identificando y extrayendo variables
- [Límites de velocidad](https://docs.claude.com/es/api/rate-limits.md): Para mitigar el mal uso y gestionar la capacidad en nuestra API, hemos implementado límites sobre cuánto puede usar una organización la API de Claude.
- [Recuperar Resultados de Lote de Mensajes](https://docs.claude.com/es/api/retrieving-message-batch-results.md): Streams the results of a Message Batch as a `.jsonl` file.
Each line in the file is a JSON object containing the result of a single request in the Message Batch. Results are not guaranteed to be in the same order as requests. Use the `custom_id` field to match results to requests.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Recuperar un Lote de Mensajes](https://docs.claude.com/es/api/retrieving-message-batches.md): This endpoint is idempotent and can be used to poll for Message Batch completion. To access the results of a Message Batch, make a request to the `results_url` field in the response.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Niveles de servicio](https://docs.claude.com/es/api/service-tiers.md): Los diferentes niveles de servicio te permiten equilibrar la disponibilidad, el rendimiento y los costos predecibles según las necesidades de tu aplicación.
- [Regiones compatibles](https://docs.claude.com/es/api/supported-regions.md): Aquí están los países, regiones y territorios desde los cuales actualmente podemos brindar acceso:
- [API de Uso y Costo](https://docs.claude.com/es/api/usage-cost-api.md): Accede programáticamente a los datos de uso y costo de la API de tu organización con la API de Administración de Uso y Costo.
- [Versiones](https://docs.claude.com/es/api/versioning.md): Al realizar solicitudes de API, debes enviar un encabezado de solicitud `anthropic-version`. Por ejemplo, `anthropic-version: 2023-06-01`. Si estás usando nuestros [SDKs de cliente](/es/api/client-sdks), esto se maneja automáticamente para ti.
- [Glosario](https://docs.claude.com/es/docs/about-claude/glossary.md): Estos conceptos no son únicos de los modelos de lenguaje de Anthropic, pero presentamos un breve resumen de términos clave a continuación.
- [Depreciaciones de modelos](https://docs.claude.com/es/docs/about-claude/model-deprecations.md)
- [Eligiendo el modelo adecuado](https://docs.claude.com/es/docs/about-claude/models/choosing-a-model.md): Seleccionar el modelo Claude óptimo para tu aplicación implica equilibrar tres consideraciones clave: capacidades, velocidad y costo. Esta guía te ayuda a tomar una decisión informada basada en tus requisitos específicos.
- [Migración a Claude 4](https://docs.claude.com/es/docs/about-claude/models/migrating-to-claude-4.md): Guía para migrar de los modelos Claude 3.7 a los modelos Claude 4.
- [Resumen de modelos](https://docs.claude.com/es/docs/about-claude/models/overview.md): Claude es una familia de modelos de lenguaje grandes de última generación desarrollados por Anthropic. Esta guía presenta nuestros modelos y compara su rendimiento con modelos heredados.
- [Precios](https://docs.claude.com/es/docs/about-claude/pricing.md): Aprende sobre la estructura de precios de Anthropic para modelos y características
- [Moderación de contenido](https://docs.claude.com/es/docs/about-claude/use-case-guides/content-moderation.md): La moderación de contenido es un aspecto crítico para mantener un entorno seguro, respetuoso y productivo en las aplicaciones digitales. En esta guía, discutiremos cómo se puede usar Claude para moderar contenido dentro de su aplicación digital.
- [Agente de soporte al cliente](https://docs.claude.com/es/docs/about-claude/use-case-guides/customer-support-chat.md): Esta guía explica cómo aprovechar las capacidades conversacionales avanzadas de Claude para manejar consultas de clientes en tiempo real, proporcionando soporte 24/7, reduciendo tiempos de espera y gestionando altos volúmenes de soporte con respuestas precisas e interacciones positivas.
- [Resumen legal](https://docs.claude.com/es/docs/about-claude/use-case-guides/legal-summarization.md): Esta guía explica cómo aprovechar las capacidades avanzadas de procesamiento de lenguaje natural de Claude para resumir eficientemente documentos legales, extrayendo información clave y acelerando la investigación legal. Con Claude, puedes optimizar la revisión de contratos, preparación de litigios y trabajo regulatorio, ahorrando tiempo y asegurando precisión en tus procesos legales.
- [Guías para casos de uso comunes](https://docs.claude.com/es/docs/about-claude/use-case-guides/overview.md)
- [Enrutamiento de tickets](https://docs.claude.com/es/docs/about-claude/use-case-guides/ticket-routing.md): Esta guía explica cómo aprovechar las capacidades avanzadas de comprensión del lenguaje natural de Claude para clasificar tickets de soporte al cliente a escala basándose en la intención del cliente, urgencia, priorización, perfil del cliente y más.
- [Complemento de Google Sheets](https://docs.claude.com/es/docs/agents-and-tools/claude-for-sheets.md): La [extensión Claude for Sheets](https://workspace.google.com/marketplace/app/claude%5Ffor%5Fsheets/909417792257) integra Claude en Google Sheets, permitiéndote ejecutar interacciones con Claude directamente en las celdas.
- [Conector MCP](https://docs.claude.com/es/docs/agents-and-tools/mcp-connector.md): Conector MCP
- [Servidores MCP remotos](https://docs.claude.com/es/docs/agents-and-tools/remote-mcp-servers.md): Servidores MCP remotos
- [Herramienta bash](https://docs.claude.com/es/docs/agents-and-tools/tool-use/bash-tool.md): La herramienta bash permite a Claude ejecutar comandos de shell en una sesión bash persistente, permitiendo operaciones del sistema, ejecución de scripts y automatización de línea de comandos.
- [Herramienta de ejecución de código](https://docs.claude.com/es/docs/agents-and-tools/tool-use/code-execution-tool.md)
- [Herramienta de uso de computadora](https://docs.claude.com/es/docs/agents-and-tools/tool-use/computer-use-tool.md): Claude puede interactuar con entornos informáticos a través de la herramienta de uso de computadora, que proporciona capacidades de captura de pantalla y control de mouse/teclado para la interacción autónoma del escritorio.
- [Streaming de herramientas de grano fino](https://docs.claude.com/es/docs/agents-and-tools/tool-use/fine-grained-tool-streaming.md): Aprende cómo usar el streaming de herramientas de grano fino para reducir la latencia al recibir parámetros grandes de herramientas.
- [Cómo implementar el uso de herramientas](https://docs.claude.com/es/docs/agents-and-tools/tool-use/implement-tool-use.md)
- [Uso de herramientas con Claude](https://docs.claude.com/es/docs/agents-and-tools/tool-use/overview.md)
- [Herramienta de editor de texto](https://docs.claude.com/es/docs/agents-and-tools/tool-use/text-editor-tool.md): Claude puede usar una herramienta de editor de texto definida por Anthropic para ver y modificar archivos de texto, ayudándote a depurar, corregir y mejorar tu código u otros documentos de texto.
- [Uso eficiente de herramientas en tokens](https://docs.claude.com/es/docs/agents-and-tools/tool-use/token-efficient-tool-use.md)
- [Herramienta de obtención web](https://docs.claude.com/es/docs/agents-and-tools/tool-use/web-fetch-tool.md)
- [Herramienta de búsqueda web](https://docs.claude.com/es/docs/agents-and-tools/tool-use/web-search-tool.md): La herramienta de búsqueda web le da a Claude acceso directo a contenido web en tiempo real, permitiéndole responder preguntas con información actualizada más allá de su fecha límite de conocimiento.
- [Procesamiento por lotes](https://docs.claude.com/es/docs/build-with-claude/batch-processing.md)
- [Citas](https://docs.claude.com/es/docs/build-with-claude/citations.md): Claude es capaz de proporcionar citas detalladas al responder preguntas sobre documentos, ayudándote a rastrear y verificar las fuentes de información en las respuestas.
- [Ventanas de contexto](https://docs.claude.com/es/docs/build-with-claude/context-windows.md)
- [Embeddings](https://docs.claude.com/es/docs/build-with-claude/embeddings.md): Los embeddings de texto son representaciones numéricas del texto que permiten medir la similitud semántica. Esta guía introduce los embeddings, sus aplicaciones y cómo usar modelos de embedding para tareas como búsqueda, recomendaciones y detección de anomalías.
- [Construyendo con pensamiento extendido](https://docs.claude.com/es/docs/build-with-claude/extended-thinking.md)
- [API de Archivos](https://docs.claude.com/es/docs/build-with-claude/files.md): La API de Archivos te permite subir y gestionar archivos para usar con la API de Claude sin tener que volver a subir contenido con cada solicitud.
- [Soporte multilingüe](https://docs.claude.com/es/docs/build-with-claude/multilingual-support.md): Claude sobresale en tareas en múltiples idiomas, manteniendo un fuerte rendimiento interlingüístico en relación con el inglés.
- [Resumen de características](https://docs.claude.com/es/docs/build-with-claude/overview.md): Explora las características y capacidades avanzadas de Claude.
- [Soporte para PDF](https://docs.claude.com/es/docs/build-with-claude/pdf-support.md): Procesa PDFs con Claude. Extrae texto, analiza gráficos y comprende contenido visual de tus documentos.
- [Caché de prompts](https://docs.claude.com/es/docs/build-with-claude/prompt-caching.md)
- [Sé claro, directo y detallado](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/be-clear-and-direct.md)
- [Deja que Claude piense (encadenamiento de pensamiento) para mejorar el rendimiento](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/chain-of-thought.md)
- [Encadena prompts complejos para un rendimiento más sólido](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/chain-prompts.md)
- [Mejores prácticas de ingeniería de prompts para Claude 4](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/claude-4-best-practices.md): Técnicas específicas de ingeniería de prompts para los modelos Claude 4 para ayudarte a lograr resultados óptimos en tus aplicaciones.
- [Consejos de pensamiento extendido](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/extended-thinking-tips.md): Estrategias avanzadas y técnicas para aprovechar al máximo las características de pensamiento extendido de Claude.
- [Consejos para el uso de contexto largo](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/long-context-tips.md)
- [Usar ejemplos (prompting multishot) para guiar el comportamiento de Claude](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/multishot-prompting.md)
- [Resumen de ingeniería de prompts](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/overview.md)
- [Prellenar la respuesta de Claude para un mayor control de la salida](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/prefill-claudes-response.md)
- [Generar automáticamente plantillas de prompts iniciales](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/prompt-generator.md)
- [Usa nuestro mejorador de prompts para optimizar tus prompts](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/prompt-improver.md)
- [Usar plantillas de prompts y variables](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/prompt-templates-and-variables.md)
- [Asignando un rol a Claude con un prompt de sistema](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/system-prompts.md)
- [Usa etiquetas XML para estructurar tus prompts](https://docs.claude.com/es/docs/build-with-claude/prompt-engineering/use-xml-tags.md)
- [Resultados de búsqueda](https://docs.claude.com/es/docs/build-with-claude/search-results.md): Habilita citas naturales para aplicaciones RAG proporcionando resultados de búsqueda con atribución de fuente
- [Transmisión de Mensajes](https://docs.claude.com/es/docs/build-with-claude/streaming.md): Aprende cómo transmitir respuestas de mensajes de forma incremental usando eventos enviados por el servidor (SSE) con la API de Claude.
- [Conteo de tokens](https://docs.claude.com/es/docs/build-with-claude/token-counting.md)
- [Visión](https://docs.claude.com/es/docs/build-with-claude/vision.md): Las familias de modelos Claude 3 y 4 vienen con nuevas capacidades de visión que permiten a Claude entender y analizar imágenes, abriendo posibilidades emocionantes para la interacción multimodal.
- [Claude Code en Amazon Bedrock](https://docs.claude.com/es/docs/claude-code/amazon-bedrock.md): Aprende sobre la configuración de Claude Code a través de Amazon Bedrock, incluyendo configuración, configuración de IAM y solución de problemas.
- [Analíticas](https://docs.claude.com/es/docs/claude-code/analytics.md): Ve información detallada de uso y métricas de productividad para el despliegue de Claude Code de tu organización.
- [Referencia de CLI](https://docs.claude.com/es/docs/claude-code/cli-reference.md): Referencia completa para la interfaz de línea de comandos de Claude Code, incluyendo comandos y banderas.
- [Flujos de trabajo comunes](https://docs.claude.com/es/docs/claude-code/common-workflows.md): Aprende sobre flujos de trabajo comunes con Claude Code.
- [Gestionar costos de manera efectiva](https://docs.claude.com/es/docs/claude-code/costs.md): Aprende cómo rastrear y optimizar el uso de tokens y costos al usar Claude Code.
- [Uso de datos](https://docs.claude.com/es/docs/claude-code/data-usage.md): Aprende sobre las políticas de uso de datos de Anthropic para Claude
- [Contenedores de desarrollo](https://docs.claude.com/es/docs/claude-code/devcontainer.md): Aprende sobre el contenedor de desarrollo de Claude Code para equipos que necesitan entornos consistentes y seguros.
- [Claude Code GitHub Actions](https://docs.claude.com/es/docs/claude-code/github-actions.md): Aprende sobre la integración de Claude Code en tu flujo de trabajo de desarrollo con Claude Code GitHub Actions
- [Claude Code GitLab CI/CD](https://docs.claude.com/es/docs/claude-code/gitlab-ci-cd.md): Aprende sobre la integración de Claude Code en tu flujo de trabajo de desarrollo con GitLab CI/CD
- [Claude Code en Google Vertex AI](https://docs.claude.com/es/docs/claude-code/google-vertex-ai.md): Aprende sobre la configuración de Claude Code a través de Google Vertex AI, incluyendo configuración, configuración de IAM y solución de problemas.
- [Referencia de hooks](https://docs.claude.com/es/docs/claude-code/hooks.md): Esta página proporciona documentación de referencia para implementar hooks en Claude Code.
- [Comienza con los hooks de Claude Code](https://docs.claude.com/es/docs/claude-code/hooks-guide.md): Aprende cómo personalizar y extender el comportamiento de Claude Code registrando comandos de shell
- [Gestión de Identidad y Acceso](https://docs.claude.com/es/docs/claude-code/iam.md): Aprende cómo configurar la autenticación de usuarios, autorización y controles de acceso para Claude Code en tu organización.
- [Agregar Claude Code a tu IDE](https://docs.claude.com/es/docs/claude-code/ide-integrations.md): Aprende cómo agregar Claude Code a tu IDE favorito
- [Modo interactivo](https://docs.claude.com/es/docs/claude-code/interactive-mode.md): Referencia completa para atajos de teclado, modos de entrada y características interactivas en las sesiones de Claude Code.
- [Legal y cumplimiento](https://docs.claude.com/es/docs/claude-code/legal-and-compliance.md): Acuerdos legales, certificaciones de cumplimiento e información de seguridad para Claude Code.
- [Configuración de gateway LLM](https://docs.claude.com/es/docs/claude-code/llm-gateway.md): Aprende cómo configurar Claude Code con soluciones de gateway LLM, incluyendo configuración de LiteLLM, métodos de autenticación y características empresariales como seguimiento de uso y gestión de presupuesto.
- [Conecta Claude Code a herramientas mediante MCP](https://docs.claude.com/es/docs/claude-code/mcp.md): Aprende cómo conectar Claude Code a tus herramientas con el Protocolo de Contexto de Modelo.
- [Gestionar la memoria de Claude](https://docs.claude.com/es/docs/claude-code/memory.md): Aprende cómo gestionar la memoria de Claude Code a través de sesiones con diferentes ubicaciones de memoria y mejores prácticas.
- [Configuración del modelo](https://docs.claude.com/es/docs/claude-code/model-config.md): Aprende sobre la configuración del modelo Claude Code, incluyendo alias de modelo como `opusplan`
- [Monitoreo](https://docs.claude.com/es/docs/claude-code/monitoring-usage.md): Aprende cómo habilitar y configurar OpenTelemetry para Claude Code.
- [Configuración de red empresarial](https://docs.claude.com/es/docs/claude-code/network-config.md): Configure Claude Code para entornos empresariales con servidores proxy, Autoridades de Certificación (CA) personalizadas y autenticación de Seguridad de Capa de Transporte mutua (mTLS).
- [Estilos de salida](https://docs.claude.com/es/docs/claude-code/output-styles.md): Adapta Claude Code para usos más allá de la ingeniería de software
- [Resumen de Claude Code](https://docs.claude.com/es/docs/claude-code/overview.md): Aprende sobre Claude Code, la herramienta de codificación agéntica de Anthropic que vive en tu terminal y te ayuda a convertir ideas en código más rápido que nunca.
- [Inicio rápido](https://docs.claude.com/es/docs/claude-code/quickstart.md): ¡Bienvenido a Claude Code!
- [Herramientas Personalizadas](https://docs.claude.com/es/docs/claude-code/sdk/custom-tools.md): Construye e integra herramientas personalizadas para extender la funcionalidad del SDK de Claude Code
- [Modificando prompts del sistema](https://docs.claude.com/es/docs/claude-code/sdk/modifying-system-prompts.md): Aprende cómo personalizar el comportamiento de Claude modificando los prompts del sistema usando tres enfoques: estilos de salida, appendSystemPrompt y customSystemPrompt.
- [Seguimiento de Costos y Uso](https://docs.claude.com/es/docs/claude-code/sdk/sdk-cost-tracking.md): Comprende y rastrea el uso de tokens para facturación en el SDK de Claude Code
- [Modo sin interfaz](https://docs.claude.com/es/docs/claude-code/sdk/sdk-headless.md): Ejecuta Claude Code programáticamente sin interfaz de usuario interactiva
- [MCP en el SDK](https://docs.claude.com/es/docs/claude-code/sdk/sdk-mcp.md): Extiende Claude Code con herramientas personalizadas usando servidores del Protocolo de Contexto de Modelo
- [Descripción general](https://docs.claude.com/es/docs/claude-code/sdk/sdk-overview.md): Construye agentes de IA personalizados con el SDK de Claude Code
- [Manejo de Permisos](https://docs.claude.com/es/docs/claude-code/sdk/sdk-permissions.md): Controla el uso de herramientas y permisos en el SDK de Claude Code
- [Referencia del SDK de Python](https://docs.claude.com/es/docs/claude-code/sdk/sdk-python.md): Referencia completa de la API para el SDK de Python de Claude Code, incluyendo todas las funciones, tipos y clases.
- [Gestión de Sesiones](https://docs.claude.com/es/docs/claude-code/sdk/sdk-sessions.md): Comprenda cómo el SDK de Claude Code maneja las sesiones y la reanudación de sesiones
- [Comandos Slash en el SDK](https://docs.claude.com/es/docs/claude-code/sdk/sdk-slash-commands.md): Aprende cómo usar comandos slash para controlar sesiones de Claude Code a través del SDK
- [Referencia del SDK de TypeScript](https://docs.claude.com/es/docs/claude-code/sdk/sdk-typescript.md): Referencia completa de la API para el SDK de TypeScript de Claude Code, incluyendo todas las funciones, tipos e interfaces.
- [Entrada de Streaming](https://docs.claude.com/es/docs/claude-code/sdk/streaming-vs-single-mode.md): Comprender los dos modos de entrada para Claude Code SDK y cuándo usar cada uno
- [Subagentes en el SDK](https://docs.claude.com/es/docs/claude-code/sdk/subagents.md): Trabajando con subagentes en el SDK de Claude Code
- [Listas de Tareas](https://docs.claude.com/es/docs/claude-code/sdk/todo-tracking.md): Rastrea y muestra tareas pendientes usando el Claude Code SDK para una gestión organizada de tareas
- [Seguridad](https://docs.claude.com/es/docs/claude-code/security.md): Aprende sobre las salvaguardas de seguridad de Claude Code y las mejores prácticas para un uso seguro.
- [Configuración de Claude Code](https://docs.claude.com/es/docs/claude-code/settings.md): Configura Claude Code con configuraciones globales y a nivel de proyecto, y variables de entorno.
- [Configurar Claude Code](https://docs.claude.com/es/docs/claude-code/setup.md): Instala, autentica y comienza a usar Claude Code en tu máquina de desarrollo.
- [Comandos slash](https://docs.claude.com/es/docs/claude-code/slash-commands.md): Controla el comportamiento de Claude durante una sesión interactiva con comandos slash.
- [Configuración de línea de estado](https://docs.claude.com/es/docs/claude-code/statusline.md): Crea una línea de estado personalizada para Claude Code para mostrar información contextual
- [Subagentes](https://docs.claude.com/es/docs/claude-code/sub-agents.md): Crea y usa subagentes de IA especializados en Claude Code para flujos de trabajo específicos de tareas y gestión mejorada del contexto.
- [Optimiza la configuración de tu terminal](https://docs.claude.com/es/docs/claude-code/terminal-config.md): Claude Code funciona mejor cuando tu terminal está configurado correctamente. Sigue estas pautas para optimizar tu experiencia.
- [Descripción general del despliegue empresarial](https://docs.claude.com/es/docs/claude-code/third-party-integrations.md): Aprende cómo Claude Code puede integrarse con varios servicios de terceros e infraestructura para cumplir con los requisitos de despliegue empresarial.
- [Solución de problemas](https://docs.claude.com/es/docs/claude-code/troubleshooting.md): Descubre soluciones a problemas comunes con la instalación y uso de Claude Code.
- [Comenzar con Claude](https://docs.claude.com/es/docs/get-started.md): Realiza tu primera llamada a la API de Claude y construye un asistente de búsqueda web simple
- [Introducción a Claude](https://docs.claude.com/es/docs/intro.md): Claude es una plataforma de IA altamente eficiente, confiable e inteligente construida por Anthropic. Claude sobresale en tareas que involucran lenguaje, razonamiento, análisis, codificación y más.
- [Política de Privacidad de Anthropic](https://docs.claude.com/es/docs/legal-center/privacy.md)
- [Seguridad y cumplimiento](https://docs.claude.com/es/docs/legal-center/security-compliance.md)
- [Construyendo con Claude](https://docs.claude.com/es/docs/overview.md): Claude es una familia de [modelos de IA altamente eficientes e inteligentes](/es/docs/about-claude/models) construidos por Anthropic. Aunque Claude es poderoso y extensible, también es la IA más confiable y fiable disponible. Sigue protocolos críticos, comete menos errores y es resistente a jailbreaks, permitiendo a [clientes empresariales](https://claude.com/customers) construir las aplicaciones impulsadas por IA más seguras a escala.
- [Define tus criterios de éxito](https://docs.claude.com/es/docs/test-and-evaluate/define-success.md)
- [Crear evaluaciones empíricas sólidas](https://docs.claude.com/es/docs/test-and-evaluate/develop-tests.md)
- [Usando la Herramienta de Evaluación](https://docs.claude.com/es/docs/test-and-evaluate/eval-tool.md): La [Consola de Claude](https://console.anthropic.com/dashboard) cuenta con una **herramienta de Evaluación** que te permite probar tus prompts bajo varios escenarios.
- [null](https://docs.claude.com/es/docs/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals.md)
- [Aumentar la consistencia de la salida (modo JSON)](https://docs.claude.com/es/docs/test-and-evaluate/strengthen-guardrails/increase-consistency.md)
- [Mantener a Claude en personaje con indicaciones de rol y precompletado](https://docs.claude.com/es/docs/test-and-evaluate/strengthen-guardrails/keep-claude-in-character.md)
- [Mitigar jailbreaks e inyecciones de prompt](https://docs.claude.com/es/docs/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks.md)
- [Reducir las alucinaciones](https://docs.claude.com/es/docs/test-and-evaluate/strengthen-guardrails/reduce-hallucinations.md)
- [Reduciendo la latencia](https://docs.claude.com/es/docs/test-and-evaluate/strengthen-guardrails/reduce-latency.md)
- [Reducir la filtración de prompts](https://docs.claude.com/es/docs/test-and-evaluate/strengthen-guardrails/reduce-prompt-leak.md)
- [null](https://docs.claude.com/es/home.md)
- [Plataforma de Desarrolladores de Claude](https://docs.claude.com/es/release-notes/api.md): Sigue las actualizaciones de la API de Claude y la Consola de Claude.
- [Aplicaciones de Claude](https://docs.claude.com/es/release-notes/claude-apps.md): Sigue las actualizaciones de las aplicaciones Claude de Anthropic.
- [Claude Code](https://docs.claude.com/es/release-notes/claude-code.md): Mantente al día con las últimas características y mejoras de Claude Code
- [Resumen](https://docs.claude.com/es/release-notes/overview.md): Sigue las actualizaciones de los productos y servicios de Anthropic.
- [Prompts del Sistema](https://docs.claude.com/es/release-notes/system-prompts.md): Ve las actualizaciones a los prompts del sistema principales en [Claude.ai](https://www.claude.ai) y las aplicaciones de Claude para [iOS](http://anthropic.com/ios) y [Android](http://anthropic.com/android).
- [Tarjeta de sistema de Claude 3.7](https://docs.claude.com/es/resources/claude-3-7-system-card.md)
- [Ficha técnica de Claude 3](https://docs.claude.com/es/resources/claude-3-model-card.md)
- [Tarjeta del Sistema Claude 4](https://docs.claude.com/es/resources/claude-4-system-card.md): Tarjeta del modelo de Anthropic para Claude Opus y Sonnet 4 con detalles de rendimiento y seguridad.
- [Cursos de Anthropic](https://docs.claude.com/es/resources/courses.md)
- [null](https://docs.claude.com/es/resources/overview.md)
- [Editor adaptativo](https://docs.claude.com/es/resources/prompt-library/adaptive-editor.md): Reescribe texto siguiendo instrucciones dadas por el usuario, como con un tono, audiencia o estilo diferente.
- [Analizador de códigos de aeropuerto](https://docs.claude.com/es/resources/prompt-library/airport-code-analyst.md): Encuentra y extrae códigos de aeropuerto del texto.
- [Antropólogo alienígena](https://docs.claude.com/es/resources/prompt-library/alien-anthropologist.md): Analiza la cultura y costumbres humanas desde la perspectiva de un antropólogo alienígena.
- [Alquimista de aliteraciones](https://docs.claude.com/es/resources/prompt-library/alliteration-alchemist.md): Genera frases y oraciones aliterativas para cualquier tema dado.
- [Transmisiones de Babel](https://docs.claude.com/es/resources/prompt-library/babels-broadcasts.md): Crea tweets convincentes de anuncios de productos en los 10 idiomas más hablados del mundo.
- [Constructor de marca](https://docs.claude.com/es/resources/prompt-library/brand-builder.md): Crea un brief de diseño para una identidad de marca holística.
- [Entrenador de carrera](https://docs.claude.com/es/resources/prompt-library/career-coach.md): Participa en conversaciones de juego de roles con un entrenador de carrera de IA.
- [Cita tus fuentes](https://docs.claude.com/es/resources/prompt-library/cite-your-sources.md): Obtén respuestas a preguntas sobre el contenido de un documento con citas relevantes que respalden la respuesta.
- [Clarificador de código](https://docs.claude.com/es/resources/prompt-library/code-clarifier.md): Simplifica y explica código complejo en lenguaje sencillo.
- [Consultor de código](https://docs.claude.com/es/resources/prompt-library/code-consultant.md): Sugiere mejoras para optimizar el rendimiento del código Python.
- [Clarividente corporativo](https://docs.claude.com/es/resources/prompt-library/corporate-clairvoyant.md): Extrae perspectivas, identifica riesgos y destila información clave de informes corporativos largos en un solo memorando.
- [Teclas Cósmicas](https://docs.claude.com/es/resources/prompt-library/cosmic-keystrokes.md): Genera un juego interactivo de mecanografía de velocidad en un solo archivo HTML, con jugabilidad de desplazamiento lateral y estilo Tailwind CSS.
- [Convertidor CSV](https://docs.claude.com/es/resources/prompt-library/csv-converter.md): Convierte datos de varios formatos (JSON, XML, etc.) en archivos CSV correctamente formateados.
- [Creador culinario](https://docs.claude.com/es/resources/prompt-library/culinary-creator.md): Sugiere ideas de recetas basadas en los ingredientes disponibles del usuario y sus preferencias dietéticas.
- [Organizador de datos](https://docs.claude.com/es/resources/prompt-library/data-organizer.md): Convierte texto no estructurado en tablas JSON personalizadas.
- [Decodificador de direcciones](https://docs.claude.com/es/resources/prompt-library/direction-decoder.md): Transforma el lenguaje natural en instrucciones paso a paso.
- [Intérprete de sueños](https://docs.claude.com/es/resources/prompt-library/dream-interpreter.md): Ofrece interpretaciones y perspectivas sobre el simbolismo de los sueños del usuario.
- [Estimador de eficiencia](https://docs.claude.com/es/resources/prompt-library/efficiency-estimator.md): Calcula la complejidad temporal de funciones y algoritmos.
- [Extractor de correos electrónicos](https://docs.claude.com/es/resources/prompt-library/email-extractor.md): Extrae direcciones de correo electrónico de un documento en una lista con formato JSON.
- [Codificador de emojis](https://docs.claude.com/es/resources/prompt-library/emoji-encoder.md): Convierte texto plano en mensajes divertidos y expresivos con emojis.
- [Navegador de dilemas éticos](https://docs.claude.com/es/resources/prompt-library/ethical-dilemma-navigator.md): Ayuda al usuario a reflexionar sobre dilemas éticos complejos y proporciona diferentes perspectivas.
- [Experto en fórmulas de Excel](https://docs.claude.com/es/resources/prompt-library/excel-formula-expert.md): Crear fórmulas de Excel basadas en cálculos o manipulaciones de datos descritos por el usuario.
- [Fabricador de funciones](https://docs.claude.com/es/resources/prompt-library/function-fabricator.md): Crear funciones de Python basadas en especificaciones detalladas.
- [Asesor de moda futurista](https://docs.claude.com/es/resources/prompt-library/futuristic-fashion-advisor.md): Sugiere tendencias y estilos de moda vanguardista para las preferencias específicas del usuario.
- [Git gud](https://docs.claude.com/es/resources/prompt-library/git-gud.md): Generar comandos Git apropiados basados en acciones de control de versiones descritas por el usuario.
- [Programador de Google Apps Script](https://docs.claude.com/es/resources/prompt-library/google-apps-scripter.md): Genera scripts de Google Apps para completar tareas basadas en los requisitos del usuario.
- [Gurú de calificación](https://docs.claude.com/es/resources/prompt-library/grading-guru.md): Compara y evalúa la calidad de textos escritos basándose en criterios y estándares definidos por el usuario.
- [Genio de la gramática](https://docs.claude.com/es/resources/prompt-library/grammar-genie.md): Transforma oraciones gramaticalmente incorrectas en inglés correcto.
- [Hal el asistente humorístico](https://docs.claude.com/es/resources/prompt-library/hal-the-humorous-helper.md): Chatea con una IA conocedora que tiene un lado sarcástico.
- [Iluminador de modismos](https://docs.claude.com/es/resources/prompt-library/idiom-illuminator.md): Explica el significado y origen de modismos y proverbios comunes.
- [Creador de preguntas de entrevista](https://docs.claude.com/es/resources/prompt-library/interview-question-crafter.md): Genera preguntas para entrevistas.
- [Leyenda de LaTeX](https://docs.claude.com/es/resources/prompt-library/latex-legend.md): Escribe documentos LaTeX, generando código para ecuaciones matemáticas, tablas y más.
- [Planificador de lecciones](https://docs.claude.com/es/resources/prompt-library/lesson-planner.md): Crea planes de lecciones detallados sobre cualquier tema.
- [Biblioteca de Prompts](https://docs.claude.com/es/resources/prompt-library/library.md)
- [Moderador maestro](https://docs.claude.com/es/resources/prompt-library/master-moderator.md): Evalúa las entradas de los usuarios en busca de contenido potencialmente dañino o ilegal.
- [Secretario de reuniones](https://docs.claude.com/es/resources/prompt-library/meeting-scribe.md): Destila reuniones en resúmenes concisos que incluyen temas de discusión, conclusiones clave y elementos de acción.
- [Maestro de memorandos](https://docs.claude.com/es/resources/prompt-library/memo-maestro.md): Redacta memorandos empresariales integrales basados en puntos clave.
- [Mentor de mindfulness](https://docs.claude.com/es/resources/prompt-library/mindfulness-mentor.md): Guía al usuario a través de ejercicios y técnicas de mindfulness para la reducción del estrés.
- [Colorizador de estados de ánimo](https://docs.claude.com/es/resources/prompt-library/mood-colorizer.md): Transforma descripciones de texto de estados de ánimo en códigos HEX correspondientes.
- [Musa motivacional](https://docs.claude.com/es/resources/prompt-library/motivational-muse.md): Proporciona mensajes motivacionales personalizados y afirmaciones basadas en la entrada del usuario.
- [Creador de neologismos](https://docs.claude.com/es/resources/prompt-library/neologism-creator.md): Inventa nuevas palabras y proporciona sus definiciones basándose en conceptos o ideas proporcionadas por el usuario.
- [Reflexionador de perspectivas](https://docs.claude.com/es/resources/prompt-library/perspectives-ponderer.md): Sopesa los pros y contras de un tema proporcionado por el usuario.
- [Reflexiones filosóficas](https://docs.claude.com/es/resources/prompt-library/philosophical-musings.md): Participa en discusiones filosóficas profundas y experimentos mentales.
- [Purificador de PII](https://docs.claude.com/es/resources/prompt-library/pii-purifier.md): Detecta y elimina automáticamente información de identificación personal (PII) de documentos de texto.
- [Superpoderes políglotas](https://docs.claude.com/es/resources/prompt-library/polyglot-superpowers.md): Traduce texto de cualquier idioma a cualquier idioma.
- [Poeta de palabras compuestas](https://docs.claude.com/es/resources/prompt-library/portmanteau-poet.md): Combina dos palabras para crear una nueva palabra compuesta significativa.
- [Experto en nombres de productos](https://docs.claude.com/es/resources/prompt-library/product-naming-pro.md): Crea nombres atractivos para productos a partir de descripciones y palabras clave.
- [Pulidor de prosa](https://docs.claude.com/es/resources/prompt-library/prose-polisher.md): Refina y mejora el contenido escrito con técnicas avanzadas de corrección de estilo y sugerencias.
- [Pun-dit](https://docs.claude.com/es/resources/prompt-library/pun-dit.md): Genera juegos de palabras ingeniosos y creativos basados en cualquier tema dado.
- [Cazador de errores de Python](https://docs.claude.com/es/resources/prompt-library/python-bug-buster.md): Detecta y corrige errores en código Python.
- [Clasificador de reseñas](https://docs.claude.com/es/resources/prompt-library/review-classifier.md): Categoriza comentarios en etiquetas y categorizaciones preespecificadas.
- [Adivina adivinanza](https://docs.claude.com/es/resources/prompt-library/riddle-me-this.md): Genera adivinanzas y guía al usuario hacia las soluciones.
- [Simulador de escenarios de ciencia ficción](https://docs.claude.com/es/resources/prompt-library/sci-fi-scenario-simulator.md): Discute con el usuario varios escenarios de ciencia ficción y los desafíos y consideraciones asociados.
- [Simplificador de segundo grado](https://docs.claude.com/es/resources/prompt-library/second-grade-simplifier.md): Haz que el texto complejo sea fácil de entender para los estudiantes jóvenes.
- [Experto en símiles](https://docs.claude.com/es/resources/prompt-library/simile-savant.md): Genera símiles a partir de descripciones básicas.
- [Sabio socrático](https://docs.claude.com/es/resources/prompt-library/socratic-sage.md): Participa en conversaciones de estilo socrático sobre un tema dado por el usuario.
- [Hechicero de hojas de cálculo](https://docs.claude.com/es/resources/prompt-library/spreadsheet-sorcerer.md): Genera hojas de cálculo CSV con varios tipos de datos.
- [Hechicero SQL](https://docs.claude.com/es/resources/prompt-library/sql-sorcerer.md): Transforma el lenguaje cotidiano en consultas SQL.
- [Compañero de narración](https://docs.claude.com/es/resources/prompt-library/storytelling-sidekick.md): Crea historias atractivas de forma colaborativa con el usuario, ofreciendo giros argumentales y desarrollo de personajes.
- [Consultor de viajes en el tiempo](https://docs.claude.com/es/resources/prompt-library/time-travel-consultant.md): Ayuda al usuario a navegar escenarios hipotéticos de viajes en el tiempo y sus implicaciones.
- [Trabalenguas](https://docs.claude.com/es/resources/prompt-library/tongue-twister.md): Crea trabalenguas desafiantes.
- [Generador de trivia](https://docs.claude.com/es/resources/prompt-library/trivia-generator.md): Genera preguntas de trivia sobre una amplia gama de temas y proporciona pistas cuando sea necesario.
- [Detector de tono de tweets](https://docs.claude.com/es/resources/prompt-library/tweet-tone-detector.md): Detecta el tono y el sentimiento detrás de los tweets.
- [Innovador de fitness VR](https://docs.claude.com/es/resources/prompt-library/vr-fitness-innovator.md): Lluvia de ideas creativas para juegos de fitness de realidad virtual.
- [Asistente de sitios web](https://docs.claude.com/es/resources/prompt-library/website-wizard.md): Crea sitios web de una sola página basados en las especificaciones del usuario.
- [Guías de inicio rápido](https://docs.claude.com/es/resources/quickstarts.md): Comienza rápidamente con Claude y la API de Claude
- [Estado del sistema](https://docs.claude.com/es/resources/status.md)
- [Obtenir la clé API](https://docs.claude.com/fr/api/admin-api/apikeys/get-api-key.md)
- [Lister les clés API](https://docs.claude.com/fr/api/admin-api/apikeys/list-api-keys.md)
- [Mettre à jour les clés API](https://docs.claude.com/fr/api/admin-api/apikeys/update-api-key.md)
- [Obtenir le rapport d'utilisation de Claude Code](https://docs.claude.com/fr/api/admin-api/claude-code/get-claude-code-usage-report.md): Retrieve daily aggregated usage metrics for Claude Code users.
Enables organizations to analyze developer productivity and build custom dashboards.
- [Créer une invitation](https://docs.claude.com/fr/api/admin-api/invites/create-invite.md)
- [Supprimer l'invitation](https://docs.claude.com/fr/api/admin-api/invites/delete-invite.md)
- [Obtenir l'invitation](https://docs.claude.com/fr/api/admin-api/invites/get-invite.md)
- [Lister les invitations](https://docs.claude.com/fr/api/admin-api/invites/list-invites.md)
- [Obtenir les informations de l'organisation](https://docs.claude.com/fr/api/admin-api/organization/get-me.md)
- [Obtenir le rapport de coûts](https://docs.claude.com/fr/api/admin-api/usage-cost/get-cost-report.md)
- [Obtenir le rapport d'utilisation pour l'API Messages](https://docs.claude.com/fr/api/admin-api/usage-cost/get-messages-usage-report.md)
- [Obtenir l'utilisateur](https://docs.claude.com/fr/api/admin-api/users/get-user.md)
- [Lister les utilisateurs](https://docs.claude.com/fr/api/admin-api/users/list-users.md)
- [Supprimer un utilisateur](https://docs.claude.com/fr/api/admin-api/users/remove-user.md)
- [Mettre à jour l'utilisateur](https://docs.claude.com/fr/api/admin-api/users/update-user.md)
- [Ajouter un membre d'espace de travail](https://docs.claude.com/fr/api/admin-api/workspace_members/create-workspace-member.md)
- [Supprimer un membre de l'espace de travail](https://docs.claude.com/fr/api/admin-api/workspace_members/delete-workspace-member.md)
- [Obtenir un membre de l'espace de travail](https://docs.claude.com/fr/api/admin-api/workspace_members/get-workspace-member.md)
- [Lister les Membres de l'Espace de Travail](https://docs.claude.com/fr/api/admin-api/workspace_members/list-workspace-members.md)
- [Mettre à jour un membre de l'espace de travail](https://docs.claude.com/fr/api/admin-api/workspace_members/update-workspace-member.md)
- [Archiver l'espace de travail](https://docs.claude.com/fr/api/admin-api/workspaces/archive-workspace.md)
- [Créer un espace de travail](https://docs.claude.com/fr/api/admin-api/workspaces/create-workspace.md)
- [Obtenir l'espace de travail](https://docs.claude.com/fr/api/admin-api/workspaces/get-workspace.md)
- [Lister les espaces de travail](https://docs.claude.com/fr/api/admin-api/workspaces/list-workspaces.md)
- [Mettre à jour l'espace de travail](https://docs.claude.com/fr/api/admin-api/workspaces/update-workspace.md)
- [Aperçu de l'API Admin](https://docs.claude.com/fr/api/administration-api.md)
- [En-têtes bêta](https://docs.claude.com/fr/api/beta-headers.md): Documentation pour l'utilisation des en-têtes bêta avec l'API Claude
- [Annuler un lot de messages](https://docs.claude.com/fr/api/canceling-message-batches.md): Batches may be canceled any time before processing ends. Once cancellation is initiated, the batch enters a `canceling` state, at which time the system may complete any in-progress, non-interruptible requests before finalizing cancellation.
The number of canceled requests is specified in `request_counts`. To determine which requests were canceled, check the individual results within the batch. Note that cancellation may not result in any canceled requests if they were non-interruptible.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [API d'Analyse de Code Claude](https://docs.claude.com/fr/api/claude-code-analytics-api.md): Accédez par programmation aux analyses d'utilisation de Claude Code et aux métriques de productivité de votre organisation avec l'API d'Administration d'Analyse de Code Claude.
- [API Amazon Bedrock](https://docs.claude.com/fr/api/claude-on-amazon-bedrock.md): Les modèles Claude d'Anthropic sont maintenant généralement disponibles via Amazon Bedrock.
- [API Vertex AI](https://docs.claude.com/fr/api/claude-on-vertex-ai.md): Les modèles Claude d'Anthropic sont maintenant généralement disponibles via [Vertex AI](https://cloud.google.com/vertex-ai).
- [SDK clients](https://docs.claude.com/fr/api/client-sdks.md): Nous fournissons des bibliothèques clients dans plusieurs langages populaires qui facilitent le travail avec l'API Claude.
- [Créer un lot de messages](https://docs.claude.com/fr/api/creating-message-batches.md): Send a batch of Message creation requests.
The Message Batches API can be used to process multiple Messages API requests at once. Once a Message Batch is created, it begins processing immediately. Batches can take up to 24 hours to complete.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Supprimer un lot de messages](https://docs.claude.com/fr/api/deleting-message-batches.md): Delete a Message Batch.
Message Batches can only be deleted once they've finished processing. If you'd like to delete an in-progress batch, you must first cancel it.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Erreurs](https://docs.claude.com/fr/api/errors.md)
- [Télécharger un fichier](https://docs.claude.com/fr/api/files-content.md): Télécharger le contenu d'un fichier généré par Claude
- [Créer un fichier](https://docs.claude.com/fr/api/files-create.md): Télécharger un fichier
- [Supprimer un fichier](https://docs.claude.com/fr/api/files-delete.md): Rendre un fichier inaccessible via l'API
- [Lister les fichiers](https://docs.claude.com/fr/api/files-list.md): Lister les fichiers dans un espace de travail
- [Obtenir les métadonnées du fichier](https://docs.claude.com/fr/api/files-metadata.md)
- [Obtenir de l'aide](https://docs.claude.com/fr/api/getting-help.md): Nous avons essayé de fournir les réponses aux questions les plus courantes dans cette documentation. Cependant, si vous avez besoin d'un support technique supplémentaire pour utiliser Claude, l'API Claude, ou l'un de nos produits, vous pouvez contacter notre équipe de support à [support.claude.com](https://support.claude.com).
- [Gestion des raisons d'arrêt](https://docs.claude.com/fr/api/handling-stop-reasons.md): Comprendre et gérer les différentes raisons d'arrêt dans les réponses de l'API Messages de Claude.
- [Adresses IP](https://docs.claude.com/fr/api/ip-addresses.md): Les services Anthropic utilisent des adresses IP fixes pour les connexions entrantes et sortantes. Vous pouvez utiliser ces adresses pour configurer vos règles de pare-feu pour un accès sécurisé à l'API Claude et à la Console. Ces adresses ne changeront pas sans préavis.
- [Lister les lots de messages](https://docs.claude.com/fr/api/listing-message-batches.md): List all Message Batches within a Workspace. Most recently created batches are returned first.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Messages](https://docs.claude.com/fr/api/messages.md): Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless multi-turn conversations.
Learn more about the Messages API in our [user guide](/en/docs/initial-setup)
- [Exemples de lots de messages](https://docs.claude.com/fr/api/messages-batch-examples.md): Exemples d'utilisation pour l'API Message Batches
- [Compter les jetons de message](https://docs.claude.com/fr/api/messages-count-tokens.md): Count the number of tokens in a Message.
The Token Count API can be used to count the number of tokens in a Message, including tools, images, and documents, without creating it.
Learn more about token counting in our [user guide](/en/docs/build-with-claude/token-counting)
- [Exemples de messages](https://docs.claude.com/fr/api/messages-examples.md): Exemples de requêtes et de réponses pour l'API Messages
- [Migration depuis les Complétions de Texte](https://docs.claude.com/fr/api/migrating-from-text-completions-to-messages.md): Migration depuis les Complétions de Texte vers les Messages
- [Obtenir un modèle](https://docs.claude.com/fr/api/models.md): Get a specific model.
The Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.
- [Lister les Modèles](https://docs.claude.com/fr/api/models-list.md): List available models.
The Models API response can be used to determine which models are available for use in the API. More recently released models are listed first.
- [Compatibilité SDK OpenAI](https://docs.claude.com/fr/api/openai-sdk.md): Anthropic fournit une couche de compatibilité qui vous permet d'utiliser le SDK OpenAI pour tester l'API Claude. Avec quelques modifications de code, vous pouvez rapidement évaluer les capacités du modèle Anthropic.
- [Aperçu](https://docs.claude.com/fr/api/overview.md): Aperçu de l'API Claude
- [Générer une invite](https://docs.claude.com/fr/api/prompt-tools-generate.md): Générer une invite bien rédigée
- [Améliorer un prompt](https://docs.claude.com/fr/api/prompt-tools-improve.md): Créer un nouveau prompt amélioré guidé par des commentaires
- [Modéliser un prompt](https://docs.claude.com/fr/api/prompt-tools-templatize.md): Modéliser un prompt en identifiant et extrayant les variables
- [Limites de taux](https://docs.claude.com/fr/api/rate-limits.md): Pour atténuer les abus et gérer la capacité de notre API, nous avons mis en place des limites sur l'utilisation que peut faire une organisation de l'API Claude.
- [Récupérer les résultats du lot de messages](https://docs.claude.com/fr/api/retrieving-message-batch-results.md): Streams the results of a Message Batch as a `.jsonl` file.
Each line in the file is a JSON object containing the result of a single request in the Message Batch. Results are not guaranteed to be in the same order as requests. Use the `custom_id` field to match results to requests.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Récupérer un lot de messages](https://docs.claude.com/fr/api/retrieving-message-batches.md): This endpoint is idempotent and can be used to poll for Message Batch completion. To access the results of a Message Batch, make a request to the `results_url` field in the response.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Niveaux de service](https://docs.claude.com/fr/api/service-tiers.md): Différents niveaux de service vous permettent d'équilibrer la disponibilité, les performances et les coûts prévisibles en fonction des besoins de votre application.
- [Régions prises en charge](https://docs.claude.com/fr/api/supported-regions.md): Voici les pays, régions et territoires depuis lesquels nous pouvons actuellement prendre en charge l'accès :
- [API d'utilisation et de coût](https://docs.claude.com/fr/api/usage-cost-api.md): Accédez par programmation aux données d'utilisation et de coût de l'API de votre organisation avec l'API d'administration d'utilisation et de coût.
- [Versions](https://docs.claude.com/fr/api/versioning.md): Lors de l'envoi de requêtes API, vous devez envoyer un en-tête de requête `anthropic-version`. Par exemple, `anthropic-version: 2023-06-01`. Si vous utilisez nos [SDK clients](/fr/api/client-sdks), ceci est géré automatiquement pour vous.
- [Glossaire](https://docs.claude.com/fr/docs/about-claude/glossary.md): Ces concepts ne sont pas uniques aux modèles de langage d'Anthropic, mais nous présentons un bref résumé des termes clés ci-dessous.
- [Dépréciations de modèles](https://docs.claude.com/fr/docs/about-claude/model-deprecations.md)
- [Choisir le bon modèle](https://docs.claude.com/fr/docs/about-claude/models/choosing-a-model.md): Sélectionner le modèle Claude optimal pour votre application implique d'équilibrer trois considérations clés : les capacités, la vitesse et le coût. Ce guide vous aide à prendre une décision éclairée en fonction de vos besoins spécifiques.
- [Migration vers Claude 4](https://docs.claude.com/fr/docs/about-claude/models/migrating-to-claude-4.md): Guide pour migrer des modèles Claude 3.7 vers les modèles Claude 4
- [Aperçu des modèles](https://docs.claude.com/fr/docs/about-claude/models/overview.md): Claude est une famille de modèles de langage de grande taille de pointe développés par Anthropic. Ce guide présente nos modèles et compare leurs performances avec les modèles hérités.
- [Tarification](https://docs.claude.com/fr/docs/about-claude/pricing.md): Découvrez la structure tarifaire d'Anthropic pour les modèles et fonctionnalités
- [Modération de contenu](https://docs.claude.com/fr/docs/about-claude/use-case-guides/content-moderation.md): La modération de contenu est un aspect critique du maintien d'un environnement sûr, respectueux et productif dans les applications numériques. Dans ce guide, nous discuterons de la façon dont Claude peut être utilisé pour modérer le contenu au sein de votre application numérique.
- [Agent de support client](https://docs.claude.com/fr/docs/about-claude/use-case-guides/customer-support-chat.md): Ce guide explique comment exploiter les capacités conversationnelles avancées de Claude pour traiter les demandes des clients en temps réel, fournir un support 24h/24 et 7j/7, réduire les temps d'attente et gérer des volumes élevés de support avec des réponses précises et des interactions positives.
- [Résumé juridique](https://docs.claude.com/fr/docs/about-claude/use-case-guides/legal-summarization.md): Ce guide explique comment exploiter les capacités avancées de traitement du langage naturel de Claude pour résumer efficacement les documents juridiques, extraire les informations clés et accélérer la recherche juridique. Avec Claude, vous pouvez rationaliser l'examen des contrats, la préparation des litiges et le travail réglementaire, économisant du temps et garantissant la précision de vos processus juridiques.
- [Guides pour les cas d'utilisation courants](https://docs.claude.com/fr/docs/about-claude/use-case-guides/overview.md)
- [Routage des tickets](https://docs.claude.com/fr/docs/about-claude/use-case-guides/ticket-routing.md): Ce guide explique comment exploiter les capacités avancées de compréhension du langage naturel de Claude pour classifier les tickets de support client à grande échelle en fonction de l'intention du client, de l'urgence, de la priorisation, du profil client, et plus encore.
- [Extension Google Sheets](https://docs.claude.com/fr/docs/agents-and-tools/claude-for-sheets.md): L'[extension Claude for Sheets](https://workspace.google.com/marketplace/app/claude%5Ffor%5Fsheets/909417792257) intègre Claude dans Google Sheets, vous permettant d'exécuter des interactions avec Claude directement dans les cellules.
- [Connecteur MCP](https://docs.claude.com/fr/docs/agents-and-tools/mcp-connector.md): Le connecteur Model Context Protocol (MCP) de Claude vous permet de vous connecter à des serveurs MCP distants directement depuis l'API Messages sans client MCP séparé.
- [Serveurs MCP distants](https://docs.claude.com/fr/docs/agents-and-tools/remote-mcp-servers.md)
- [Outil bash](https://docs.claude.com/fr/docs/agents-and-tools/tool-use/bash-tool.md): L'outil bash permet à Claude d'exécuter des commandes shell dans une session bash persistante, permettant les opérations système, l'exécution de scripts et l'automatisation en ligne de commande.
- [Outil d'exécution de code](https://docs.claude.com/fr/docs/agents-and-tools/tool-use/code-execution-tool.md)
- [Outil d'utilisation d'ordinateur](https://docs.claude.com/fr/docs/agents-and-tools/tool-use/computer-use-tool.md): Claude peut interagir avec les environnements informatiques grâce à l'outil d'utilisation d'ordinateur, qui fournit des capacités de capture d'écran et un contrôle souris/clavier pour l'interaction autonome du bureau.
- [Streaming d'outils à granularité fine](https://docs.claude.com/fr/docs/agents-and-tools/tool-use/fine-grained-tool-streaming.md): Streaming d'outils à granularité fine
- [Comment implémenter l'utilisation d'outils](https://docs.claude.com/fr/docs/agents-and-tools/tool-use/implement-tool-use.md)
- [Utilisation d'outils avec Claude](https://docs.claude.com/fr/docs/agents-and-tools/tool-use/overview.md)
- [Outil éditeur de texte](https://docs.claude.com/fr/docs/agents-and-tools/tool-use/text-editor-tool.md): Claude peut utiliser un outil éditeur de texte défini par Anthropic pour visualiser et modifier des fichiers texte, vous aidant à déboguer, corriger et améliorer votre code ou autres documents texte.
- [Utilisation d'outils économe en jetons](https://docs.claude.com/fr/docs/agents-and-tools/tool-use/token-efficient-tool-use.md)
- [Outil de récupération web](https://docs.claude.com/fr/docs/agents-and-tools/tool-use/web-fetch-tool.md)
- [Outil de recherche web](https://docs.claude.com/fr/docs/agents-and-tools/tool-use/web-search-tool.md): L'outil de recherche web donne à Claude un accès direct au contenu web en temps réel, lui permettant de répondre aux questions avec des informations à jour au-delà de sa limite de connaissances.
- [Traitement par lots](https://docs.claude.com/fr/docs/build-with-claude/batch-processing.md)
- [Citations](https://docs.claude.com/fr/docs/build-with-claude/citations.md): Claude est capable de fournir des citations détaillées lors de la réponse à des questions sur des documents, vous aidant à suivre et vérifier les sources d'information dans les réponses.
- [Fenêtres de contexte](https://docs.claude.com/fr/docs/build-with-claude/context-windows.md): Comprendre les fenêtres de contexte de Claude et comment elles fonctionnent avec la réflexion étendue et l'utilisation d'outils.
- [Embeddings](https://docs.claude.com/fr/docs/build-with-claude/embeddings.md): Les embeddings de texte sont des représentations numériques de texte qui permettent de mesurer la similarité sémantique. Ce guide présente les embeddings, leurs applications, et comment utiliser les modèles d'embedding pour des tâches comme la recherche, les recommandations, et la détection d'anomalies.
- [Construire avec la pensée étendue](https://docs.claude.com/fr/docs/build-with-claude/extended-thinking.md)
- [API Files](https://docs.claude.com/fr/docs/build-with-claude/files.md)
- [Support multilingue](https://docs.claude.com/fr/docs/build-with-claude/multilingual-support.md): Claude excelle dans les tâches impliquant plusieurs langues, maintenant de solides performances interlinguistiques par rapport à l'anglais.
- [Aperçu des fonctionnalités](https://docs.claude.com/fr/docs/build-with-claude/overview.md): Explorez les fonctionnalités et capacités avancées de Claude.
- [Support PDF](https://docs.claude.com/fr/docs/build-with-claude/pdf-support.md): Traitez les PDF avec Claude. Extrayez du texte, analysez des graphiques et comprenez le contenu visuel de vos documents.
- [Mise en cache des prompts](https://docs.claude.com/fr/docs/build-with-claude/prompt-caching.md)
- [Soyez clair, direct et détaillé](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/be-clear-and-direct.md)
- [Laissez Claude réfléchir (prompting par chaîne de pensée) pour améliorer ses performances](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/chain-of-thought.md)
- [Chaîner des prompts complexes pour de meilleures performances](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/chain-prompts.md)
- [Meilleures pratiques d'ingénierie de prompts pour Claude 4](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/claude-4-best-practices.md): Guide spécialisé pour optimiser l'utilisation des modèles Claude 4 avec des techniques d'ingénierie de prompts efficaces.
- [Conseils pour la réflexion étendue](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/extended-thinking-tips.md)
- [Conseils pour les prompts avec contexte long](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/long-context-tips.md)
- [Utiliser des exemples (prompting multishot) pour guider le comportement de Claude](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/multishot-prompting.md)
- [Aperçu de l'ingénierie de prompts](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/overview.md)
- [Préremplir la réponse de Claude pour un meilleur contrôle des résultats](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/prefill-claudes-response.md)
- [Générer automatiquement des modèles de prompts en première version](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/prompt-generator.md)
- [Utilisez notre améliorateur de prompts pour optimiser vos prompts](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/prompt-improver.md)
- [Utiliser des modèles de prompts et des variables](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/prompt-templates-and-variables.md)
- [Donner un rôle à Claude avec une invite système](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/system-prompts.md)
- [Utiliser les balises XML pour structurer vos prompts](https://docs.claude.com/fr/docs/build-with-claude/prompt-engineering/use-xml-tags.md)
- [Résultats de recherche](https://docs.claude.com/fr/docs/build-with-claude/search-results.md): Activez les citations naturelles pour les applications RAG en fournissant des résultats de recherche avec attribution de source
- [Messages en streaming](https://docs.claude.com/fr/docs/build-with-claude/streaming.md): Apprenez à diffuser des réponses de messages de manière incrémentielle en utilisant des événements envoyés par le serveur (SSE).
- [Comptage de tokens](https://docs.claude.com/fr/docs/build-with-claude/token-counting.md)
- [Vision](https://docs.claude.com/fr/docs/build-with-claude/vision.md): Les familles de modèles Claude 3 et 4 sont dotées de nouvelles capacités de vision qui permettent à Claude de comprendre et d'analyser les images, ouvrant des possibilités passionnantes pour l'interaction multimodale.
- [Claude Code sur Amazon Bedrock](https://docs.claude.com/fr/docs/claude-code/amazon-bedrock.md): Apprenez à configurer Claude Code via Amazon Bedrock, y compris la configuration, la configuration IAM et le dépannage.
- [Analytiques](https://docs.claude.com/fr/docs/claude-code/analytics.md): Consultez des informations détaillées sur l'utilisation et les métriques de productivité pour le déploiement Claude Code de votre organisation.
- [Référence CLI](https://docs.claude.com/fr/docs/claude-code/cli-reference.md): Référence complète pour l'interface en ligne de commande Claude Code, incluant les commandes et les drapeaux.
- [Flux de travail courants](https://docs.claude.com/fr/docs/claude-code/common-workflows.md): Apprenez les flux de travail courants avec Claude Code.
- [Gérer les coûts efficacement](https://docs.claude.com/fr/docs/claude-code/costs.md): Apprenez à suivre et optimiser l'utilisation des jetons et les coûts lors de l'utilisation de Claude Code.
- [Utilisation des données](https://docs.claude.com/fr/docs/claude-code/data-usage.md): Découvrez les politiques d'utilisation des données d'Anthropic pour Claude
- [Conteneurs de développement](https://docs.claude.com/fr/docs/claude-code/devcontainer.md): Découvrez le conteneur de développement Claude Code pour les équipes qui ont besoin d'environnements cohérents et sécurisés.
- [Claude Code GitHub Actions](https://docs.claude.com/fr/docs/claude-code/github-actions.md): Apprenez à intégrer Claude Code dans votre flux de travail de développement avec Claude Code GitHub Actions
- [Claude Code GitLab CI/CD](https://docs.claude.com/fr/docs/claude-code/gitlab-ci-cd.md): Apprenez à intégrer Claude Code dans votre flux de travail de développement avec GitLab CI/CD
- [Claude Code sur Google Vertex AI](https://docs.claude.com/fr/docs/claude-code/google-vertex-ai.md): Apprenez à configurer Claude Code via Google Vertex AI, y compris la configuration, la configuration IAM et le dépannage.
- [Référence des hooks](https://docs.claude.com/fr/docs/claude-code/hooks.md): Cette page fournit une documentation de référence pour l'implémentation des hooks dans Claude Code.
- [Commencer avec les hooks Claude Code](https://docs.claude.com/fr/docs/claude-code/hooks-guide.md): Apprenez à personnaliser et étendre le comportement de Claude Code en enregistrant des commandes shell
- [Gestion des identités et des accès](https://docs.claude.com/fr/docs/claude-code/iam.md): Apprenez à configurer l'authentification des utilisateurs, l'autorisation et les contrôles d'accès pour Claude Code dans votre organisation.
- [Ajouter Claude Code à votre IDE](https://docs.claude.com/fr/docs/claude-code/ide-integrations.md): Apprenez comment ajouter Claude Code à votre IDE préféré
- [Mode interactif](https://docs.claude.com/fr/docs/claude-code/interactive-mode.md): Référence complète pour les raccourcis clavier, les modes de saisie et les fonctionnalités interactives dans les sessions Claude Code.
- [Juridique et conformité](https://docs.claude.com/fr/docs/claude-code/legal-and-compliance.md): Accords juridiques, certifications de conformité et informations de sécurité pour Claude Code.
- [Configuration de la passerelle LLM](https://docs.claude.com/fr/docs/claude-code/llm-gateway.md): Apprenez à configurer Claude Code avec des solutions de passerelle LLM, y compris la configuration LiteLLM, les méthodes d'authentification et les fonctionnalités d'entreprise comme le suivi d'utilisation et la gestion budgétaire.
- [Connecter Claude Code aux outils via MCP](https://docs.claude.com/fr/docs/claude-code/mcp.md): Apprenez à connecter Claude Code à vos outils avec le Model Context Protocol.
- [Gérer la mémoire de Claude](https://docs.claude.com/fr/docs/claude-code/memory.md): Apprenez à gérer la mémoire de Claude Code à travers les sessions avec différents emplacements de mémoire et les meilleures pratiques.
- [Configuration du modèle](https://docs.claude.com/fr/docs/claude-code/model-config.md): Apprenez la configuration du modèle Claude Code, y compris les alias de modèles comme `opusplan`
- [Surveillance](https://docs.claude.com/fr/docs/claude-code/monitoring-usage.md): Apprenez comment activer et configurer OpenTelemetry pour Claude Code.
- [Configuration réseau d'entreprise](https://docs.claude.com/fr/docs/claude-code/network-config.md): Configurez Claude Code pour les environnements d'entreprise avec des serveurs proxy, des Autorités de Certification (CA) personnalisées et l'authentification Transport Layer Security mutuelle (mTLS).
- [Styles de sortie](https://docs.claude.com/fr/docs/claude-code/output-styles.md): Adaptez Claude Code pour des utilisations au-delà de l'ingénierie logicielle
- [Aperçu de Claude Code](https://docs.claude.com/fr/docs/claude-code/overview.md): Découvrez Claude Code, l'outil de codage agentique d'Anthropic qui vit dans votre terminal et vous aide à transformer vos idées en code plus rapidement que jamais.
- [Démarrage rapide](https://docs.claude.com/fr/docs/claude-code/quickstart.md): Bienvenue dans Claude Code !
- [Outils Personnalisés](https://docs.claude.com/fr/docs/claude-code/sdk/custom-tools.md): Créez et intégrez des outils personnalisés pour étendre les fonctionnalités du SDK Claude Code
- [Modification des invites système](https://docs.claude.com/fr/docs/claude-code/sdk/modifying-system-prompts.md): Apprenez à personnaliser le comportement de Claude en modifiant les invites système en utilisant trois approches - styles de sortie, appendSystemPrompt, et customSystemPrompt.
- [Suivi des Coûts et de l'Utilisation](https://docs.claude.com/fr/docs/claude-code/sdk/sdk-cost-tracking.md): Comprendre et suivre l'utilisation des tokens pour la facturation dans le SDK Claude Code
- [Mode sans interface](https://docs.claude.com/fr/docs/claude-code/sdk/sdk-headless.md): Exécuter Claude Code de manière programmatique sans interface utilisateur interactive
- [MCP dans le SDK](https://docs.claude.com/fr/docs/claude-code/sdk/sdk-mcp.md): Étendez Claude Code avec des outils personnalisés en utilisant les serveurs Model Context Protocol
- [Aperçu](https://docs.claude.com/fr/docs/claude-code/sdk/sdk-overview.md): Construisez des agents IA personnalisés avec le SDK Claude Code
- [Gestion des Permissions](https://docs.claude.com/fr/docs/claude-code/sdk/sdk-permissions.md): Contrôlez l'utilisation des outils et les permissions dans le SDK Claude Code
- [Référence du SDK Python](https://docs.claude.com/fr/docs/claude-code/sdk/sdk-python.md): Référence API complète pour le SDK Python Claude Code, incluant toutes les fonctions, types et classes.
- [Gestion des Sessions](https://docs.claude.com/fr/docs/claude-code/sdk/sdk-sessions.md): Comprendre comment le SDK Claude Code gère les sessions et la reprise de session
- [Commandes Slash dans le SDK](https://docs.claude.com/fr/docs/claude-code/sdk/sdk-slash-commands.md): Apprenez à utiliser les commandes slash pour contrôler les sessions Claude Code via le SDK
- [Référence du SDK TypeScript](https://docs.claude.com/fr/docs/claude-code/sdk/sdk-typescript.md): Référence API complète pour le SDK TypeScript Claude Code, incluant toutes les fonctions, types et interfaces.
- [Entrée en Streaming](https://docs.claude.com/fr/docs/claude-code/sdk/streaming-vs-single-mode.md): Comprendre les deux modes d'entrée pour le SDK Claude Code et quand utiliser chacun
- [Sous-agents dans le SDK](https://docs.claude.com/fr/docs/claude-code/sdk/subagents.md): Travailler avec les sous-agents dans le SDK Claude Code
- [Listes de tâches](https://docs.claude.com/fr/docs/claude-code/sdk/todo-tracking.md): Suivez et affichez les tâches à faire en utilisant le SDK Claude Code pour une gestion organisée des tâches
- [Sécurité](https://docs.claude.com/fr/docs/claude-code/security.md): Découvrez les mesures de sécurité de Claude Code et les meilleures pratiques pour une utilisation sûre.
- [Paramètres de Claude Code](https://docs.claude.com/fr/docs/claude-code/settings.md): Configurez Claude Code avec des paramètres globaux et au niveau du projet, ainsi que des variables d'environnement.
- [Configurer Claude Code](https://docs.claude.com/fr/docs/claude-code/setup.md): Installez, authentifiez et commencez à utiliser Claude Code sur votre machine de développement.
- [Commandes slash](https://docs.claude.com/fr/docs/claude-code/slash-commands.md): Contrôlez le comportement de Claude pendant une session interactive avec les commandes slash.
- [Configuration de la ligne de statut](https://docs.claude.com/fr/docs/claude-code/statusline.md): Créez une ligne de statut personnalisée pour Claude Code afin d'afficher des informations contextuelles
- [Sous-agents](https://docs.claude.com/fr/docs/claude-code/sub-agents.md): Créez et utilisez des sous-agents IA spécialisés dans Claude Code pour des flux de travail spécifiques aux tâches et une gestion améliorée du contexte.
- [Optimisez votre configuration de terminal](https://docs.claude.com/fr/docs/claude-code/terminal-config.md): Claude Code fonctionne mieux lorsque votre terminal est correctement configuré. Suivez ces directives pour optimiser votre expérience.
- [Aperçu du déploiement d'entreprise](https://docs.claude.com/fr/docs/claude-code/third-party-integrations.md): Découvrez comment Claude Code peut s'intégrer avec divers services tiers et infrastructures pour répondre aux exigences de déploiement d'entreprise.
- [Dépannage](https://docs.claude.com/fr/docs/claude-code/troubleshooting.md): Découvrez des solutions aux problèmes courants d'installation et d'utilisation de Claude Code.
- [Commencer avec Claude](https://docs.claude.com/fr/docs/get-started.md): Effectuez votre premier appel API vers Claude et créez un assistant de recherche web simple
- [Introduction à Claude](https://docs.claude.com/fr/docs/intro.md): Claude est une plateforme d'IA hautement performante, fiable et intelligente construite par Anthropic. Claude excelle dans les tâches impliquant le langage, le raisonnement, l'analyse, le codage, et plus encore.
- [Politique de Confidentialité d'Anthropic](https://docs.claude.com/fr/docs/legal-center/privacy.md)
- [Sécurité et conformité](https://docs.claude.com/fr/docs/legal-center/security-compliance.md)
- [Construire avec Claude](https://docs.claude.com/fr/docs/overview.md): Claude est une famille de [modèles d'IA hautement performants et intelligents](/fr/docs/about-claude/models) construits par Anthropic. Bien que Claude soit puissant et extensible, c'est aussi l'IA la plus fiable et digne de confiance disponible. Il suit des protocoles critiques, fait moins d'erreurs et résiste aux jailbreaks—permettant aux [clients d'entreprise](https://claude.com/customers) de construire les applications alimentées par l'IA les plus sûres à grande échelle.
- [Définir vos critères de réussite](https://docs.claude.com/fr/docs/test-and-evaluate/define-success.md)
- [Créer des évaluations empiriques solides](https://docs.claude.com/fr/docs/test-and-evaluate/develop-tests.md)
- [Utilisation de l'outil d'évaluation](https://docs.claude.com/fr/docs/test-and-evaluate/eval-tool.md): La [Console Claude](https://console.anthropic.com/dashboard) dispose d'un **outil d'évaluation** qui vous permet de tester vos prompts sous différents scénarios.
- [null](https://docs.claude.com/fr/docs/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals.md)
- [Améliorer la cohérence des résultats (mode JSON)](https://docs.claude.com/fr/docs/test-and-evaluate/strengthen-guardrails/increase-consistency.md)
- [Garder Claude dans son rôle avec le prompt de rôle et le pré-remplissage](https://docs.claude.com/fr/docs/test-and-evaluate/strengthen-guardrails/keep-claude-in-character.md)
- [Atténuer les jailbreaks et les injections de prompt](https://docs.claude.com/fr/docs/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks.md)
- [Réduire les hallucinations](https://docs.claude.com/fr/docs/test-and-evaluate/strengthen-guardrails/reduce-hallucinations.md)
- [Réduction de la latence](https://docs.claude.com/fr/docs/test-and-evaluate/strengthen-guardrails/reduce-latency.md)
- [Réduire la fuite de prompt](https://docs.claude.com/fr/docs/test-and-evaluate/strengthen-guardrails/reduce-prompt-leak.md)
- [null](https://docs.claude.com/fr/home.md)
- [Plateforme de développement Claude](https://docs.claude.com/fr/release-notes/api.md): Suivez les mises à jour de l'API Claude et de la Console Claude.
- [Applications Claude](https://docs.claude.com/fr/release-notes/claude-apps.md): Suivez les mises à jour des applications Claude d'Anthropic.
- [Claude Code](https://docs.claude.com/fr/release-notes/claude-code.md): Restez à jour avec les dernières fonctionnalités et améliorations de Claude Code
- [Aperçu](https://docs.claude.com/fr/release-notes/overview.md): Suivez les mises à jour des produits et services d'Anthropic.
- [Prompts Système](https://docs.claude.com/fr/release-notes/system-prompts.md): Voir les mises à jour des prompts système principaux sur [Claude.ai](https://www.claude.ai) et les applications Claude [iOS](http://anthropic.com/ios) et [Android](http://anthropic.com/android).
- [Fiche système de Claude 3.7](https://docs.claude.com/fr/resources/claude-3-7-system-card.md)
- [Fiche technique du modèle Claude 3](https://docs.claude.com/fr/resources/claude-3-model-card.md)
- [Fiche Système Claude 4](https://docs.claude.com/fr/resources/claude-4-system-card.md)
- [Cours Anthropic](https://docs.claude.com/fr/resources/courses.md)
- [null](https://docs.claude.com/fr/resources/overview.md)
- [Éditeur adaptatif](https://docs.claude.com/fr/resources/prompt-library/adaptive-editor.md): Réécrire du texte en suivant les instructions données par l'utilisateur, comme avec un ton, un public ou un style différent.
- [Analyseur de codes d'aéroport](https://docs.claude.com/fr/resources/prompt-library/airport-code-analyst.md): Trouvez et extrayez les codes d'aéroport du texte.
- [Anthropologue extraterrestre](https://docs.claude.com/fr/resources/prompt-library/alien-anthropologist.md): Analysez la culture et les coutumes humaines du point de vue d'un anthropologue extraterrestre.
- [Alchimiste d'allitération](https://docs.claude.com/fr/resources/prompt-library/alliteration-alchemist.md): Générez des phrases et des expressions allitératives pour n'importe quel sujet donné.
- [Les diffusions de Babel](https://docs.claude.com/fr/resources/prompt-library/babels-broadcasts.md): Créez des tweets d'annonce de produit convaincants dans les 10 langues les plus parlées au monde.
- [Créateur de marque](https://docs.claude.com/fr/resources/prompt-library/brand-builder.md): Élaborez un brief de conception pour une identité de marque holistique.
- [Coach de carrière](https://docs.claude.com/fr/resources/prompt-library/career-coach.md): Participez à des conversations de jeu de rôle avec un coach de carrière IA.
- [Citez vos sources](https://docs.claude.com/fr/resources/prompt-library/cite-your-sources.md): Obtenez des réponses aux questions sur le contenu d'un document avec des citations pertinentes soutenant la réponse.
- [Clarificateur de code](https://docs.claude.com/fr/resources/prompt-library/code-clarifier.md): Simplifiez et expliquez du code complexe en langage simple.
- [Consultant en code](https://docs.claude.com/fr/resources/prompt-library/code-consultant.md): Suggérer des améliorations pour optimiser les performances du code Python.
- [Clairvoyant d'entreprise](https://docs.claude.com/fr/resources/prompt-library/corporate-clairvoyant.md): Extraire des insights, identifier les risques et distiller les informations clés de longs rapports d'entreprise en un seul mémo.
- [Frappes Cosmiques](https://docs.claude.com/fr/resources/prompt-library/cosmic-keystrokes.md): Générez un jeu de frappe rapide interactif dans un seul fichier HTML, avec un gameplay à défilement latéral et un style Tailwind CSS.
- [Convertisseur CSV](https://docs.claude.com/fr/resources/prompt-library/csv-converter.md): Convertir des données de différents formats (JSON, XML, etc.) en fichiers CSV correctement formatés.
- [Créateur culinaire](https://docs.claude.com/fr/resources/prompt-library/culinary-creator.md): Suggérer des idées de recettes basées sur les ingrédients disponibles de l'utilisateur et ses préférences alimentaires.
- [Organisateur de données](https://docs.claude.com/fr/resources/prompt-library/data-organizer.md): Transformez du texte non structuré en tableaux JSON personnalisés.
- [Décodeur de directions](https://docs.claude.com/fr/resources/prompt-library/direction-decoder.md): Transformez le langage naturel en instructions étape par étape.
- [Interprète de rêves](https://docs.claude.com/fr/resources/prompt-library/dream-interpreter.md): Offrir des interprétations et des perspectives sur le symbolisme des rêves de l'utilisateur.
- [Estimateur d'efficacité](https://docs.claude.com/fr/resources/prompt-library/efficiency-estimator.md): Calculez la complexité temporelle des fonctions et algorithmes.
- [Extracteur d'e-mails](https://docs.claude.com/fr/resources/prompt-library/email-extractor.md): Extraire les adresses e-mail d'un document dans une liste formatée JSON.
- [Encodeur d'emoji](https://docs.claude.com/fr/resources/prompt-library/emoji-encoder.md): Convertir du texte brut en messages emoji amusants et expressifs.
- [Navigateur de dilemme éthique](https://docs.claude.com/fr/resources/prompt-library/ethical-dilemma-navigator.md): Aidez l'utilisateur à réfléchir à des dilemmes éthiques complexes et fournissez différentes perspectives.
- [Expert en formules Excel](https://docs.claude.com/fr/resources/prompt-library/excel-formula-expert.md): Créer des formules Excel basées sur des calculs ou des manipulations de données décrits par l'utilisateur.
- [Fabricant de fonctions](https://docs.claude.com/fr/resources/prompt-library/function-fabricator.md): Créer des fonctions Python basées sur des spécifications détaillées.
- [Conseiller en mode futuriste](https://docs.claude.com/fr/resources/prompt-library/futuristic-fashion-advisor.md): Suggérer des tendances et styles de mode avant-gardistes pour les préférences spécifiques de l'utilisateur.
- [Git gud](https://docs.claude.com/fr/resources/prompt-library/git-gud.md): Générer des commandes Git appropriées basées sur les actions de contrôle de version décrites par l'utilisateur.
- [Scripteur d'applications Google](https://docs.claude.com/fr/resources/prompt-library/google-apps-scripter.md): Générer des scripts Google Apps pour accomplir des tâches basées sur les exigences de l'utilisateur.
- [Gourou de la notation](https://docs.claude.com/fr/resources/prompt-library/grading-guru.md): Comparez et évaluez la qualité des textes écrits selon des critères et des normes définis par l'utilisateur.
- [Génie de la grammaire](https://docs.claude.com/fr/resources/prompt-library/grammar-genie.md): Transformez des phrases grammaticalement incorrectes en anglais correct.
- [Hal l'assistant humoristique](https://docs.claude.com/fr/resources/prompt-library/hal-the-humorous-helper.md): Discutez avec une IA compétente qui a un côté sarcastique.
- [Éclaireur d'expressions idiomatiques](https://docs.claude.com/fr/resources/prompt-library/idiom-illuminator.md): Expliquer la signification et l'origine d'expressions idiomatiques et de proverbes courants.
- [Créateur de questions d'entretien](https://docs.claude.com/fr/resources/prompt-library/interview-question-crafter.md): Générer des questions pour les entretiens.
- [Légende LaTeX](https://docs.claude.com/fr/resources/prompt-library/latex-legend.md): Rédigez des documents LaTeX en générant du code pour les équations mathématiques, les tableaux et plus encore.
- [Planificateur de cours](https://docs.claude.com/fr/resources/prompt-library/lesson-planner.md): Créez des plans de cours approfondis sur n'importe quel sujet.
- [Bibliothèque de prompts](https://docs.claude.com/fr/resources/prompt-library/library.md)
- [Modérateur maître](https://docs.claude.com/fr/resources/prompt-library/master-moderator.md): Évaluez les entrées des utilisateurs pour détecter un contenu potentiellement nuisible ou illégal.
- [Secrétaire de réunion](https://docs.claude.com/fr/resources/prompt-library/meeting-scribe.md): Distillez les réunions en résumés concis incluant les sujets de discussion, les points clés à retenir et les éléments d'action.
- [Maître des mémos](https://docs.claude.com/fr/resources/prompt-library/memo-maestro.md): Rédigez des mémos d'entreprise complets basés sur des points clés.
- [Mentor de pleine conscience](https://docs.claude.com/fr/resources/prompt-library/mindfulness-mentor.md): Guidez l'utilisateur à travers des exercices et techniques de pleine conscience pour la réduction du stress.
- [Colorisateur d'humeur](https://docs.claude.com/fr/resources/prompt-library/mood-colorizer.md): Transformez les descriptions textuelles d'humeurs en codes HEX correspondants.
- [Muse motivationnelle](https://docs.claude.com/fr/resources/prompt-library/motivational-muse.md): Fournir des messages motivationnels personnalisés et des affirmations basés sur les contributions de l'utilisateur.
- [Créateur de néologismes](https://docs.claude.com/fr/resources/prompt-library/neologism-creator.md): Inventez de nouveaux mots et fournissez leurs définitions basées sur des concepts ou idées fournis par l'utilisateur.
- [Réflexion sur les perspectives](https://docs.claude.com/fr/resources/prompt-library/perspectives-ponderer.md): Pesez le pour et le contre d'un sujet fourni par l'utilisateur.
- [Réflexions philosophiques](https://docs.claude.com/fr/resources/prompt-library/philosophical-musings.md): Engagez-vous dans des discussions philosophiques profondes et des expériences de pensée.
- [Purificateur d'informations personnelles](https://docs.claude.com/fr/resources/prompt-library/pii-purifier.md): Détecte et supprime automatiquement les informations personnellement identifiables (PII) des documents texte.
- [Superpouvoirs polyglottes](https://docs.claude.com/fr/resources/prompt-library/polyglot-superpowers.md): Traduisez du texte de n'importe quelle langue vers n'importe quelle langue.
- [Poète de mots-valises](https://docs.claude.com/fr/resources/prompt-library/portmanteau-poet.md): Mélangez deux mots ensemble pour créer un nouveau mot-valise significatif.
- [Pro du naming de produits](https://docs.claude.com/fr/resources/prompt-library/product-naming-pro.md): Créez des noms de produits accrocheurs à partir de descriptions et de mots-clés.
- [Polisseur de prose](https://docs.claude.com/fr/resources/prompt-library/prose-polisher.md): Affinez et améliorez le contenu écrit avec des techniques de révision avancées et des suggestions.
- [Pun-dit](https://docs.claude.com/fr/resources/prompt-library/pun-dit.md): Générer des jeux de mots astucieux et des calembours basés sur n'importe quel sujet donné.
- [Chasseur de bugs Python](https://docs.claude.com/fr/resources/prompt-library/python-bug-buster.md): Détecter et corriger les bugs dans le code Python.
- [Classificateur d'avis](https://docs.claude.com/fr/resources/prompt-library/review-classifier.md): Catégorisez les commentaires dans des balises et catégorisations pré-spécifiées.
- [Devine-moi ça](https://docs.claude.com/fr/resources/prompt-library/riddle-me-this.md): Générer des énigmes et guider l'utilisateur vers les solutions.
- [Simulateur de scénarios de science-fiction](https://docs.claude.com/fr/resources/prompt-library/sci-fi-scenario-simulator.md): Discutez avec l'utilisateur de divers scénarios de science-fiction et des défis et considérations associés.
- [Simplificateur de niveau CE1](https://docs.claude.com/fr/resources/prompt-library/second-grade-simplifier.md): Rendre les textes complexes faciles à comprendre pour les jeunes apprenants.
- [Expert en comparaisons](https://docs.claude.com/fr/resources/prompt-library/simile-savant.md): Générer des comparaisons à partir de descriptions de base.
- [Sage socratique](https://docs.claude.com/fr/resources/prompt-library/socratic-sage.md): Engagez-vous dans une conversation de style socratique sur un sujet donné par l'utilisateur.
- [Sorcier de feuilles de calcul](https://docs.claude.com/fr/resources/prompt-library/spreadsheet-sorcerer.md): Générer des feuilles de calcul CSV avec différents types de données.
- [Sorcier SQL](https://docs.claude.com/fr/resources/prompt-library/sql-sorcerer.md): Transformez le langage courant en requêtes SQL.
- [Acolyte de narration](https://docs.claude.com/fr/resources/prompt-library/storytelling-sidekick.md): Créez collaborativement des histoires captivantes avec l'utilisateur, en offrant des rebondissements et un développement de personnages.
- [Consultant en voyage dans le temps](https://docs.claude.com/fr/resources/prompt-library/time-travel-consultant.md): Aidez l'utilisateur à naviguer dans des scénarios hypothétiques de voyage dans le temps et leurs implications.
- [Virelangue](https://docs.claude.com/fr/resources/prompt-library/tongue-twister.md): Créer des virelangues difficiles.
- [Générateur de quiz](https://docs.claude.com/fr/resources/prompt-library/trivia-generator.md): Générez des questions de quiz sur une large gamme de sujets et fournissez des indices lorsque nécessaire.
- [Détecteur de ton de tweet](https://docs.claude.com/fr/resources/prompt-library/tweet-tone-detector.md): Détecter le ton et le sentiment derrière les tweets.
- [Innovateur de fitness VR](https://docs.claude.com/fr/resources/prompt-library/vr-fitness-innovator.md): Générez des idées créatives pour des jeux de fitness en réalité virtuelle.
- [Assistant de création de sites web](https://docs.claude.com/fr/resources/prompt-library/website-wizard.md): Créer des sites web d'une page basés sur les spécifications de l'utilisateur.
- [Démarrages rapides](https://docs.claude.com/fr/resources/quickstarts.md): Commencez rapidement avec Claude et l'API Claude
- [Statut du système](https://docs.claude.com/fr/resources/status.md)
- [Dapatkan Kunci API](https://docs.claude.com/id/api/admin-api/apikeys/get-api-key.md)
- [Daftar Kunci API](https://docs.claude.com/id/api/admin-api/apikeys/list-api-keys.md)
- [Perbarui Kunci API](https://docs.claude.com/id/api/admin-api/apikeys/update-api-key.md)
- [Dapatkan Laporan Penggunaan Claude Code](https://docs.claude.com/id/api/admin-api/claude-code/get-claude-code-usage-report.md): Retrieve daily aggregated usage metrics for Claude Code users.
Enables organizations to analyze developer productivity and build custom dashboards.
- [Buat Undangan](https://docs.claude.com/id/api/admin-api/invites/create-invite.md)
- [Hapus Undangan](https://docs.claude.com/id/api/admin-api/invites/delete-invite.md)
- [Dapatkan Undangan](https://docs.claude.com/id/api/admin-api/invites/get-invite.md)
- [Daftar Undangan](https://docs.claude.com/id/api/admin-api/invites/list-invites.md)
- [Dapatkan Informasi Organisasi](https://docs.claude.com/id/api/admin-api/organization/get-me.md)
- [Dapatkan Laporan Biaya](https://docs.claude.com/id/api/admin-api/usage-cost/get-cost-report.md)
- [Dapatkan Laporan Penggunaan untuk API Messages](https://docs.claude.com/id/api/admin-api/usage-cost/get-messages-usage-report.md)
- [Dapatkan Pengguna](https://docs.claude.com/id/api/admin-api/users/get-user.md)
- [Daftar Pengguna](https://docs.claude.com/id/api/admin-api/users/list-users.md)
- [Hapus Pengguna](https://docs.claude.com/id/api/admin-api/users/remove-user.md)
- [Perbarui Pengguna](https://docs.claude.com/id/api/admin-api/users/update-user.md)
- [Tambah Anggota Workspace](https://docs.claude.com/id/api/admin-api/workspace_members/create-workspace-member.md)
- [Hapus Anggota Workspace](https://docs.claude.com/id/api/admin-api/workspace_members/delete-workspace-member.md)
- [Dapatkan Anggota Workspace](https://docs.claude.com/id/api/admin-api/workspace_members/get-workspace-member.md)
- [Daftar Anggota Workspace](https://docs.claude.com/id/api/admin-api/workspace_members/list-workspace-members.md)
- [Perbarui Anggota Workspace](https://docs.claude.com/id/api/admin-api/workspace_members/update-workspace-member.md)
- [Arsipkan Ruang Kerja](https://docs.claude.com/id/api/admin-api/workspaces/archive-workspace.md)
- [Buat Workspace](https://docs.claude.com/id/api/admin-api/workspaces/create-workspace.md)
- [Dapatkan Workspace](https://docs.claude.com/id/api/admin-api/workspaces/get-workspace.md)
- [Daftar Ruang Kerja](https://docs.claude.com/id/api/admin-api/workspaces/list-workspaces.md)
- [Perbarui Workspace](https://docs.claude.com/id/api/admin-api/workspaces/update-workspace.md)
- [Ikhtisar Admin API](https://docs.claude.com/id/api/administration-api.md)
- [Header beta](https://docs.claude.com/id/api/beta-headers.md): Dokumentasi untuk menggunakan header beta dengan API Claude
- [Batalkan Batch Pesan](https://docs.claude.com/id/api/canceling-message-batches.md): Batches may be canceled any time before processing ends. Once cancellation is initiated, the batch enters a `canceling` state, at which time the system may complete any in-progress, non-interruptible requests before finalizing cancellation.
The number of canceled requests is specified in `request_counts`. To determine which requests were canceled, check the individual results within the batch. Note that cancellation may not result in any canceled requests if they were non-interruptible.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Claude Code Analytics API](https://docs.claude.com/id/api/claude-code-analytics-api.md): Akses analitik penggunaan Claude Code organisasi Anda secara terprogram dan metrik produktivitas dengan Claude Code Analytics Admin API.
- [Amazon Bedrock API](https://docs.claude.com/id/api/claude-on-amazon-bedrock.md): Model Claude dari Anthropic sekarang tersedia secara umum melalui Amazon Bedrock.
- [Vertex AI API](https://docs.claude.com/id/api/claude-on-vertex-ai.md): Model Claude dari Anthropic sekarang tersedia secara umum melalui [Vertex AI](https://cloud.google.com/vertex-ai).
- [SDK Klien](https://docs.claude.com/id/api/client-sdks.md): Kami menyediakan pustaka klien dalam sejumlah bahasa populer yang memudahkan bekerja dengan API Claude.
- [Buat Batch Pesan](https://docs.claude.com/id/api/creating-message-batches.md): Send a batch of Message creation requests.
The Message Batches API can be used to process multiple Messages API requests at once. Once a Message Batch is created, it begins processing immediately. Batches can take up to 24 hours to complete.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Hapus Batch Pesan](https://docs.claude.com/id/api/deleting-message-batches.md): Delete a Message Batch.
Message Batches can only be deleted once they've finished processing. If you'd like to delete an in-progress batch, you must first cancel it.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Kesalahan](https://docs.claude.com/id/api/errors.md): Panduan lengkap tentang kode kesalahan HTTP, batas ukuran permintaan, bentuk kesalahan, dan penanganan permintaan panjang dalam API Anthropic.
- [Unduh File](https://docs.claude.com/id/api/files-content.md): Unduh konten file yang dihasilkan Claude
- [Buat File](https://docs.claude.com/id/api/files-create.md): Unggah file
- [Hapus File](https://docs.claude.com/id/api/files-delete.md): Membuat file tidak dapat diakses melalui API
- [Daftar File](https://docs.claude.com/id/api/files-list.md): Daftar file dalam workspace
- [Dapatkan Metadata File](https://docs.claude.com/id/api/files-metadata.md)
- [Mendapatkan bantuan](https://docs.claude.com/id/api/getting-help.md): Kami telah berusaha memberikan jawaban untuk pertanyaan yang paling umum dalam dokumentasi ini. Namun, jika Anda memerlukan dukungan teknis lebih lanjut dalam menggunakan Claude, Claude API, atau produk kami lainnya, Anda dapat menghubungi tim dukungan kami di [support.claude.com](https://support.claude.com).
- [Menangani alasan berhenti](https://docs.claude.com/id/api/handling-stop-reasons.md): Pelajari cara menangani berbagai nilai stop_reason dalam respons API Messages Claude untuk membangun aplikasi yang lebih robust.
- [Alamat IP](https://docs.claude.com/id/api/ip-addresses.md): Layanan Anthropic menggunakan alamat IP tetap untuk koneksi masuk dan keluar. Anda dapat menggunakan alamat-alamat ini untuk mengonfigurasi aturan firewall Anda untuk akses aman ke Claude API dan Console. Alamat-alamat ini tidak akan berubah tanpa pemberitahuan.
- [Daftar Batch Pesan](https://docs.claude.com/id/api/listing-message-batches.md): List all Message Batches within a Workspace. Most recently created batches are returned first.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Pesan](https://docs.claude.com/id/api/messages.md): Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless multi-turn conversations.
Learn more about the Messages API in our [user guide](/en/docs/initial-setup)
- [Contoh Message Batches](https://docs.claude.com/id/api/messages-batch-examples.md): Contoh penggunaan untuk Message Batches API
- [Hitung Token Pesan](https://docs.claude.com/id/api/messages-count-tokens.md): Count the number of tokens in a Message.
The Token Count API can be used to count the number of tokens in a Message, including tools, images, and documents, without creating it.
Learn more about token counting in our [user guide](/en/docs/build-with-claude/token-counting)
- [Contoh messages](https://docs.claude.com/id/api/messages-examples.md): Contoh permintaan dan respons untuk Messages API
- [Migrasi dari Text Completions](https://docs.claude.com/id/api/migrating-from-text-completions-to-messages.md): Migrasi dari Text Completions ke Messages
- [Dapatkan Model](https://docs.claude.com/id/api/models.md): Get a specific model.
The Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.
- [Daftar Model](https://docs.claude.com/id/api/models-list.md): List available models.
The Models API response can be used to determine which models are available for use in the API. More recently released models are listed first.
- [Kompatibilitas SDK OpenAI](https://docs.claude.com/id/api/openai-sdk.md): Anthropic menyediakan lapisan kompatibilitas yang memungkinkan Anda menggunakan SDK OpenAI untuk menguji Claude API. Dengan beberapa perubahan kode, Anda dapat dengan cepat mengevaluasi kemampuan model Anthropic.
- [Ikhtisar](https://docs.claude.com/id/api/overview.md): Ikhtisar API Claude
- [Menghasilkan prompt](https://docs.claude.com/id/api/prompt-tools-generate.md): Menghasilkan prompt yang ditulis dengan baik
- [Meningkatkan prompt](https://docs.claude.com/id/api/prompt-tools-improve.md): Membuat prompt baru yang lebih baik dengan panduan umpan balik
- [Membuat template prompt](https://docs.claude.com/id/api/prompt-tools-templatize.md): Membuat template prompt dengan mengidentifikasi dan mengekstrak variabel
- [Batas tarif](https://docs.claude.com/id/api/rate-limits.md): Untuk mengurangi penyalahgunaan dan mengelola kapasitas pada API kami, kami telah menerapkan batas pada seberapa banyak organisasi dapat menggunakan Claude API.
- [Ambil Hasil Batch Pesan](https://docs.claude.com/id/api/retrieving-message-batch-results.md): Streams the results of a Message Batch as a `.jsonl` file.
Each line in the file is a JSON object containing the result of a single request in the Message Batch. Results are not guaranteed to be in the same order as requests. Use the `custom_id` field to match results to requests.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Mengambil Batch Pesan](https://docs.claude.com/id/api/retrieving-message-batches.md): This endpoint is idempotent and can be used to poll for Message Batch completion. To access the results of a Message Batch, make a request to the `results_url` field in the response.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Tingkatan layanan](https://docs.claude.com/id/api/service-tiers.md): Tingkatan layanan yang berbeda memungkinkan Anda menyeimbangkan ketersediaan, performa, dan biaya yang dapat diprediksi berdasarkan kebutuhan aplikasi Anda.
- [Wilayah yang didukung](https://docs.claude.com/id/api/supported-regions.md): Berikut adalah negara, wilayah, dan teritorial yang saat ini dapat kami dukung aksesnya:
- [API Penggunaan dan Biaya](https://docs.claude.com/id/api/usage-cost-api.md): Akses data penggunaan dan biaya API organisasi Anda secara programatis dengan API Admin Penggunaan & Biaya.
- [Versi](https://docs.claude.com/id/api/versioning.md): Saat membuat permintaan API, Anda harus mengirim header permintaan `anthropic-version`. Misalnya, `anthropic-version: 2023-06-01`. Jika Anda menggunakan [SDK klien](/id/api/client-sdks) kami, ini ditangani untuk Anda secara otomatis.
- [Glosarium](https://docs.claude.com/id/docs/about-claude/glossary.md): Konsep-konsep ini tidak unik untuk model bahasa Anthropic, tetapi kami menyajikan ringkasan singkat tentang istilah-istilah kunci di bawah ini.
- [Penghentian model](https://docs.claude.com/id/docs/about-claude/model-deprecations.md)
- [Memilih model yang tepat](https://docs.claude.com/id/docs/about-claude/models/choosing-a-model.md): Memilih model Claude yang optimal untuk aplikasi Anda melibatkan penyeimbangan tiga pertimbangan utama: kemampuan, kecepatan, dan biaya. Panduan ini membantu Anda membuat keputusan berdasarkan informasi sesuai dengan kebutuhan spesifik Anda.
- [Migrasi ke Claude 4](https://docs.claude.com/id/docs/about-claude/models/migrating-to-claude-4.md): Panduan untuk bermigrasi dari model Claude 3.7 ke model Claude 4.
- [Ikhtisar model](https://docs.claude.com/id/docs/about-claude/models/overview.md): Claude adalah keluarga model bahasa besar canggih yang dikembangkan oleh Anthropic. Panduan ini memperkenalkan model kami dan membandingkan kinerja mereka dengan model lama.
- [Harga](https://docs.claude.com/id/docs/about-claude/pricing.md): Pelajari struktur harga Anthropic untuk model dan fitur
- [Moderasi konten](https://docs.claude.com/id/docs/about-claude/use-case-guides/content-moderation.md): Moderasi konten adalah aspek penting dalam mempertahankan lingkungan yang aman, menghormati, dan produktif dalam aplikasi digital. Dalam panduan ini, kami akan membahas bagaimana Claude dapat digunakan untuk memoderasi konten dalam aplikasi digital Anda.
- [Agen dukungan pelanggan](https://docs.claude.com/id/docs/about-claude/use-case-guides/customer-support-chat.md): Panduan ini menjelaskan cara memanfaatkan kemampuan percakapan canggih Claude untuk menangani pertanyaan pelanggan secara real-time, menyediakan dukungan 24/7, mengurangi waktu tunggu, dan mengelola volume dukungan tinggi dengan respons yang akurat dan interaksi yang positif.
- [Ringkasan hukum](https://docs.claude.com/id/docs/about-claude/use-case-guides/legal-summarization.md): Panduan ini menjelaskan cara memanfaatkan kemampuan pemrosesan bahasa alami canggih Claude untuk meringkas dokumen hukum secara efisien, mengekstrak informasi kunci dan mempercepat penelitian hukum. Dengan Claude, Anda dapat menyederhanakan peninjauan kontrak, persiapan litigasi, dan pekerjaan regulasi, menghemat waktu dan memastikan akurasi dalam proses hukum Anda.
- [Panduan untuk kasus penggunaan umum](https://docs.claude.com/id/docs/about-claude/use-case-guides/overview.md)
- [Perutean tiket](https://docs.claude.com/id/docs/about-claude/use-case-guides/ticket-routing.md): Panduan ini menjelaskan cara memanfaatkan kemampuan pemahaman bahasa alami canggih Claude untuk mengklasifikasikan tiket dukungan pelanggan secara besar-besaran berdasarkan niat pelanggan, urgensi, prioritas, profil pelanggan, dan lainnya.
- [Add-on Google Sheets](https://docs.claude.com/id/docs/agents-and-tools/claude-for-sheets.md): [Ekstensi Claude for Sheets](https://workspace.google.com/marketplace/app/claude%5Ffor%5Fsheets/909417792257) mengintegrasikan Claude ke dalam Google Sheets, memungkinkan Anda menjalankan interaksi dengan Claude langsung di dalam sel.
- [Konektor MCP](https://docs.claude.com/id/docs/agents-and-tools/mcp-connector.md): Fitur konektor Model Context Protocol (MCP) Claude memungkinkan Anda terhubung ke server MCP jarak jauh langsung dari Messages API tanpa klien MCP terpisah.
- [Server MCP jarak jauh](https://docs.claude.com/id/docs/agents-and-tools/remote-mcp-servers.md): Server MCP jarak jauh yang dapat dihubungkan oleh pengembang melalui API konektor MCP Anthropic.
- [Tool bash](https://docs.claude.com/id/docs/agents-and-tools/tool-use/bash-tool.md): Tool bash memungkinkan Claude untuk menjalankan perintah shell dalam sesi bash yang persisten, memungkinkan operasi sistem, eksekusi skrip, dan otomasi command-line.
- [Tool eksekusi kode](https://docs.claude.com/id/docs/agents-and-tools/tool-use/code-execution-tool.md)
- [Tool penggunaan komputer](https://docs.claude.com/id/docs/agents-and-tools/tool-use/computer-use-tool.md): Claude dapat berinteraksi dengan lingkungan komputer melalui tool penggunaan komputer, yang menyediakan kemampuan screenshot dan kontrol mouse/keyboard untuk interaksi desktop otonom.
- [Streaming tool berbutir halus](https://docs.claude.com/id/docs/agents-and-tools/tool-use/fine-grained-tool-streaming.md): Tool use sekarang mendukung streaming berbutir halus untuk nilai parameter, memungkinkan developer untuk streaming parameter tool use tanpa buffering/validasi JSON.
- [Cara mengimplementasikan penggunaan tool](https://docs.claude.com/id/docs/agents-and-tools/tool-use/implement-tool-use.md): Panduan lengkap untuk mengimplementasikan penggunaan tool dengan Claude, termasuk definisi tool, kontrol output, dan penanganan error.
- [Penggunaan tool dengan Claude](https://docs.claude.com/id/docs/agents-and-tools/tool-use/overview.md)
- [Alat editor teks](https://docs.claude.com/id/docs/agents-and-tools/tool-use/text-editor-tool.md): Claude dapat menggunakan alat editor teks yang didefinisikan Anthropic untuk melihat dan memodifikasi file teks, membantu Anda debug, memperbaiki, dan meningkatkan kode atau dokumen teks lainnya.
- [Penggunaan tool yang efisien token](https://docs.claude.com/id/docs/agents-and-tools/tool-use/token-efficient-tool-use.md)
- [Alat web fetch](https://docs.claude.com/id/docs/agents-and-tools/tool-use/web-fetch-tool.md): Alat web fetch memungkinkan Claude mengambil konten lengkap dari halaman web dan dokumen PDF yang ditentukan.
- [Alat pencarian web](https://docs.claude.com/id/docs/agents-and-tools/tool-use/web-search-tool.md): Alat pencarian web memberikan Claude akses langsung ke konten web real-time, memungkinkannya menjawab pertanyaan dengan informasi terkini melampaui batas pengetahuannya.
- [Pemrosesan batch](https://docs.claude.com/id/docs/build-with-claude/batch-processing.md)
- [Kutipan](https://docs.claude.com/id/docs/build-with-claude/citations.md): Claude mampu memberikan kutipan terperinci saat menjawab pertanyaan tentang dokumen, membantu Anda melacak dan memverifikasi sumber informasi dalam respons.
- [Jendela konteks](https://docs.claude.com/id/docs/build-with-claude/context-windows.md)
- [Embeddings](https://docs.claude.com/id/docs/build-with-claude/embeddings.md): Embeddings teks adalah representasi numerik dari teks yang memungkinkan pengukuran kesamaan semantik. Panduan ini memperkenalkan embeddings, aplikasinya, dan cara menggunakan model embedding untuk tugas seperti pencarian, rekomendasi, dan deteksi anomali.
- [Membangun dengan extended thinking](https://docs.claude.com/id/docs/build-with-claude/extended-thinking.md)
- [Files API](https://docs.claude.com/id/docs/build-with-claude/files.md): API Files memungkinkan Anda mengunggah dan mengelola file untuk digunakan dengan Claude API tanpa perlu mengunggah ulang konten pada setiap permintaan.
- [Dukungan multibahasa](https://docs.claude.com/id/docs/build-with-claude/multilingual-support.md): Claude unggul dalam tugas-tugas lintas berbagai bahasa, mempertahankan performa lintas bahasa yang kuat relatif terhadap bahasa Inggris.
- [Gambaran umum fitur](https://docs.claude.com/id/docs/build-with-claude/overview.md): Jelajahi fitur-fitur canggih dan kemampuan Claude.
- [Dukungan PDF](https://docs.claude.com/id/docs/build-with-claude/pdf-support.md): Proses PDF dengan Claude. Ekstrak teks, analisis grafik, dan pahami konten visual dari dokumen Anda.
- [Prompt caching](https://docs.claude.com/id/docs/build-with-claude/prompt-caching.md): Prompt caching adalah fitur yang mengoptimalkan penggunaan API Anda dengan memungkinkan melanjutkan dari prefiks tertentu dalam prompt Anda.
- [Bersikaplah jelas, langsung, dan terperinci](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/be-clear-and-direct.md)
- [Biarkan Claude berpikir (chain of thought prompting) untuk meningkatkan performa](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/chain-of-thought.md)
- [Merangkai prompt kompleks untuk performa yang lebih kuat](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/chain-prompts.md)
- [Praktik terbaik rekayasa prompt Claude 4](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/claude-4-best-practices.md): Panduan ini menyediakan teknik rekayasa prompt khusus untuk model Claude 4 guna membantu Anda mencapai hasil optimal dalam aplikasi Anda.
- [Tips pemikiran diperpanjang](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/extended-thinking-tips.md): Panduan lanjutan untuk memaksimalkan fitur pemikiran diperpanjang Claude dengan strategi dan teknik yang efektif.
- [Tips prompting untuk konteks panjang](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/long-context-tips.md)
- [Gunakan contoh (multishot prompting) untuk memandu perilaku Claude](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/multishot-prompting.md)
- [Ikhtisar rekayasa prompt](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/overview.md): Panduan komprehensif untuk rekayasa prompt dengan Claude, mencakup teknik dari dasar hingga lanjutan untuk mengoptimalkan kinerja model AI.
- [Mengisi awal respons Claude untuk kontrol output yang lebih baik](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/prefill-claudes-response.md)
- [Secara otomatis menghasilkan template prompt draf pertama](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/prompt-generator.md)
- [Gunakan perbaikan prompt kami untuk mengoptimalkan prompt Anda](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/prompt-improver.md): Gunakan perbaikan prompt kami untuk mengoptimalkan prompt Anda
- [Gunakan template prompt dan variabel](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/prompt-templates-and-variables.md)
- [Memberikan Claude sebuah peran dengan prompt sistem](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/system-prompts.md)
- [Gunakan tag XML untuk menyusun prompt Anda](https://docs.claude.com/id/docs/build-with-claude/prompt-engineering/use-xml-tags.md)
- [Hasil pencarian](https://docs.claude.com/id/docs/build-with-claude/search-results.md): Aktifkan kutipan alami untuk aplikasi RAG dengan menyediakan hasil pencarian dengan atribusi sumber
- [Streaming Messages](https://docs.claude.com/id/docs/build-with-claude/streaming.md): Pelajari cara menggunakan streaming untuk menerima respons secara bertahap menggunakan server-sent events (SSE).
- [Penghitungan token](https://docs.claude.com/id/docs/build-with-claude/token-counting.md)
- [Vision](https://docs.claude.com/id/docs/build-with-claude/vision.md): Keluarga model Claude 3 dan 4 hadir dengan kemampuan vision baru yang memungkinkan Claude memahami dan menganalisis gambar, membuka kemungkinan menarik untuk interaksi multimodal.
- [Claude Code di Amazon Bedrock](https://docs.claude.com/id/docs/claude-code/amazon-bedrock.md): Pelajari tentang mengonfigurasi Claude Code melalui Amazon Bedrock, termasuk pengaturan, konfigurasi IAM, dan pemecahan masalah.
- [Analitik](https://docs.claude.com/id/docs/claude-code/analytics.md): Lihat wawasan penggunaan terperinci dan metrik produktivitas untuk deployment Claude Code organisasi Anda.
- [Referensi CLI](https://docs.claude.com/id/docs/claude-code/cli-reference.md): Referensi lengkap untuk antarmuka baris perintah Claude Code, termasuk perintah dan flag.
- [Alur kerja umum](https://docs.claude.com/id/docs/claude-code/common-workflows.md): Pelajari tentang alur kerja umum dengan Claude Code.
- [Kelola biaya secara efektif](https://docs.claude.com/id/docs/claude-code/costs.md): Pelajari cara melacak dan mengoptimalkan penggunaan token dan biaya saat menggunakan Claude Code.
- [Penggunaan data](https://docs.claude.com/id/docs/claude-code/data-usage.md): Pelajari tentang kebijakan penggunaan data Anthropic untuk Claude
- [Kontainer pengembangan](https://docs.claude.com/id/docs/claude-code/devcontainer.md): Pelajari tentang kontainer pengembangan Claude Code untuk tim yang membutuhkan lingkungan yang konsisten dan aman.
- [Claude Code GitHub Actions](https://docs.claude.com/id/docs/claude-code/github-actions.md): Pelajari tentang mengintegrasikan Claude Code ke dalam alur kerja pengembangan Anda dengan Claude Code GitHub Actions
- [Claude Code GitLab CI/CD](https://docs.claude.com/id/docs/claude-code/gitlab-ci-cd.md): Pelajari tentang mengintegrasikan Claude Code ke dalam alur kerja pengembangan Anda dengan GitLab CI/CD
- [Claude Code di Google Vertex AI](https://docs.claude.com/id/docs/claude-code/google-vertex-ai.md): Pelajari tentang mengonfigurasi Claude Code melalui Google Vertex AI, termasuk pengaturan, konfigurasi IAM, dan pemecahan masalah.
- [Referensi hooks](https://docs.claude.com/id/docs/claude-code/hooks.md): Halaman ini menyediakan dokumentasi referensi untuk mengimplementasikan hooks di Claude Code.
- [Memulai dengan Claude Code hooks](https://docs.claude.com/id/docs/claude-code/hooks-guide.md): Pelajari cara menyesuaikan dan memperluas perilaku Claude Code dengan mendaftarkan perintah shell
- [Manajemen Identitas dan Akses](https://docs.claude.com/id/docs/claude-code/iam.md): Pelajari cara mengonfigurasi autentikasi pengguna, otorisasi, dan kontrol akses untuk Claude Code di organisasi Anda.
- [Tambahkan Claude Code ke IDE Anda](https://docs.claude.com/id/docs/claude-code/ide-integrations.md): Pelajari cara menambahkan Claude Code ke IDE favorit Anda
- [Mode interaktif](https://docs.claude.com/id/docs/claude-code/interactive-mode.md): Referensi lengkap untuk pintasan keyboard, mode input, dan fitur interaktif dalam sesi Claude Code.
- [Hukum dan kepatuhan](https://docs.claude.com/id/docs/claude-code/legal-and-compliance.md): Perjanjian hukum, sertifikasi kepatuhan, dan informasi keamanan untuk Claude Code.
- [Konfigurasi gateway LLM](https://docs.claude.com/id/docs/claude-code/llm-gateway.md): Pelajari cara mengonfigurasi Claude Code dengan solusi gateway LLM, termasuk pengaturan LiteLLM, metode autentikasi, dan fitur enterprise seperti pelacakan penggunaan dan manajemen anggaran.
- [Hubungkan Claude Code ke alat melalui MCP](https://docs.claude.com/id/docs/claude-code/mcp.md): Pelajari cara menghubungkan Claude Code ke alat Anda dengan Model Context Protocol.
- [Kelola memori Claude](https://docs.claude.com/id/docs/claude-code/memory.md): Pelajari cara mengelola memori Claude Code di berbagai sesi dengan lokasi memori yang berbeda dan praktik terbaik.
- [Konfigurasi model](https://docs.claude.com/id/docs/claude-code/model-config.md): Pelajari tentang konfigurasi model Claude Code, termasuk alias model seperti `opusplan`
- [Pemantauan](https://docs.claude.com/id/docs/claude-code/monitoring-usage.md): Pelajari cara mengaktifkan dan mengonfigurasi OpenTelemetry untuk Claude Code.
- [Konfigurasi jaringan enterprise](https://docs.claude.com/id/docs/claude-code/network-config.md): Konfigurasikan Claude Code untuk lingkungan enterprise dengan server proxy, Certificate Authority (CA) kustom, dan autentikasi mutual Transport Layer Security (mTLS).
- [Gaya output](https://docs.claude.com/id/docs/claude-code/output-styles.md): Adaptasi Claude Code untuk penggunaan di luar rekayasa perangkat lunak
- [Ikhtisar Claude Code](https://docs.claude.com/id/docs/claude-code/overview.md): Pelajari tentang Claude Code, alat coding agentic dari Anthropic yang berada di terminal Anda dan membantu Anda mengubah ide menjadi kode lebih cepat dari sebelumnya.
- [Panduan Cepat](https://docs.claude.com/id/docs/claude-code/quickstart.md): Selamat datang di Claude Code!
- [Alat Kustom](https://docs.claude.com/id/docs/claude-code/sdk/custom-tools.md): Bangun dan integrasikan alat kustom untuk memperluas fungsionalitas Claude Code SDK
- [Memodifikasi prompt sistem](https://docs.claude.com/id/docs/claude-code/sdk/modifying-system-prompts.md): Pelajari cara menyesuaikan perilaku Claude dengan memodifikasi prompt sistem menggunakan tiga pendekatan - gaya output, appendSystemPrompt, dan customSystemPrompt.
- [Melacak Biaya dan Penggunaan](https://docs.claude.com/id/docs/claude-code/sdk/sdk-cost-tracking.md): Memahami dan melacak penggunaan token untuk penagihan dalam Claude Code SDK
- [Mode headless](https://docs.claude.com/id/docs/claude-code/sdk/sdk-headless.md): Jalankan Claude Code secara programatis tanpa UI interaktif
- [MCP di SDK](https://docs.claude.com/id/docs/claude-code/sdk/sdk-mcp.md): Perluas Claude Code dengan alat kustom menggunakan server Model Context Protocol
- [Ikhtisar](https://docs.claude.com/id/docs/claude-code/sdk/sdk-overview.md): Bangun agen AI kustom dengan Claude Code SDK
- [Menangani Izin](https://docs.claude.com/id/docs/claude-code/sdk/sdk-permissions.md): Kontrol penggunaan alat dan izin dalam Claude Code SDK
- [Referensi SDK Python](https://docs.claude.com/id/docs/claude-code/sdk/sdk-python.md): Referensi API lengkap untuk Claude Code Python SDK, termasuk semua fungsi, tipe, dan kelas.
- [Manajemen Sesi](https://docs.claude.com/id/docs/claude-code/sdk/sdk-sessions.md): Memahami bagaimana Claude Code SDK menangani sesi dan pemulihan sesi
- [Perintah Slash dalam SDK](https://docs.claude.com/id/docs/claude-code/sdk/sdk-slash-commands.md): Pelajari cara menggunakan perintah slash untuk mengontrol sesi Claude Code melalui SDK
- [Referensi TypeScript SDK](https://docs.claude.com/id/docs/claude-code/sdk/sdk-typescript.md): Referensi API lengkap untuk Claude Code TypeScript SDK, termasuk semua fungsi, tipe, dan interface.
- [Input Streaming](https://docs.claude.com/id/docs/claude-code/sdk/streaming-vs-single-mode.md): Memahami dua mode input untuk Claude Code SDK dan kapan menggunakan masing-masing
- [Subagen dalam SDK](https://docs.claude.com/id/docs/claude-code/sdk/subagents.md): Bekerja dengan subagen dalam Claude Code SDK
- [Daftar Todo](https://docs.claude.com/id/docs/claude-code/sdk/todo-tracking.md): Lacak dan tampilkan todo menggunakan Claude Code SDK untuk manajemen tugas yang terorganisir
- [Keamanan](https://docs.claude.com/id/docs/claude-code/security.md): Pelajari tentang perlindungan keamanan Claude Code dan praktik terbaik untuk penggunaan yang aman.
- [Pengaturan Claude Code](https://docs.claude.com/id/docs/claude-code/settings.md): Konfigurasikan Claude Code dengan pengaturan global dan tingkat proyek, serta variabel lingkungan.
- [Siapkan Claude Code](https://docs.claude.com/id/docs/claude-code/setup.md): Instal, autentikasi, dan mulai menggunakan Claude Code di mesin pengembangan Anda.
- [Perintah slash](https://docs.claude.com/id/docs/claude-code/slash-commands.md): Kontrol perilaku Claude selama sesi interaktif dengan perintah slash.
- [Konfigurasi baris status](https://docs.claude.com/id/docs/claude-code/statusline.md): Buat baris status kustom untuk Claude Code untuk menampilkan informasi kontekstual
- [Subagen](https://docs.claude.com/id/docs/claude-code/sub-agents.md): Buat dan gunakan subagen AI khusus di Claude Code untuk alur kerja spesifik tugas dan manajemen konteks yang lebih baik.
- [Optimalkan pengaturan terminal Anda](https://docs.claude.com/id/docs/claude-code/terminal-config.md): Claude Code bekerja paling baik ketika terminal Anda dikonfigurasi dengan benar. Ikuti panduan ini untuk mengoptimalkan pengalaman Anda.
- [Ikhtisar deployment enterprise](https://docs.claude.com/id/docs/claude-code/third-party-integrations.md): Pelajari bagaimana Claude Code dapat terintegrasi dengan berbagai layanan pihak ketiga dan infrastruktur untuk memenuhi persyaratan deployment enterprise.
- [Pemecahan Masalah](https://docs.claude.com/id/docs/claude-code/troubleshooting.md): Temukan solusi untuk masalah umum dengan instalasi dan penggunaan Claude Code.
- [Memulai dengan Claude](https://docs.claude.com/id/docs/get-started.md): Buat panggilan API pertama Anda ke Claude dan bangun asisten pencarian web sederhana
- [Pengantar Claude](https://docs.claude.com/id/docs/intro.md): Claude adalah platform AI yang sangat berkinerja tinggi, terpercaya, dan cerdas yang dibangun oleh Anthropic. Claude unggul dalam tugas-tugas yang melibatkan bahasa, penalaran, analisis, coding, dan lainnya.
- [Kebijakan Privasi Anthropic](https://docs.claude.com/id/docs/legal-center/privacy.md)
- [Keamanan dan kepatuhan](https://docs.claude.com/id/docs/legal-center/security-compliance.md)
- [Membangun dengan Claude](https://docs.claude.com/id/docs/overview.md): Claude adalah keluarga [model AI yang sangat berkinerja tinggi dan cerdas](/id/docs/about-claude/models) yang dibangun oleh Anthropic. Meskipun Claude kuat dan dapat diperluas, ia juga merupakan AI yang paling dapat dipercaya dan andal yang tersedia. Ia mengikuti protokol kritis, membuat lebih sedikit kesalahan, dan tahan terhadap jailbreak—memungkinkan [pelanggan enterprise](https://claude.com/customers) untuk membangun aplikasi bertenaga AI yang paling aman dalam skala besar.
- [Tentukan kriteria keberhasilan Anda](https://docs.claude.com/id/docs/test-and-evaluate/define-success.md)
- [Buat evaluasi empiris yang kuat](https://docs.claude.com/id/docs/test-and-evaluate/develop-tests.md): Pelajari cara mengembangkan kasus uji yang efektif untuk mengukur kinerja LLM terhadap kriteria kesuksesan Anda.
- [Menggunakan Alat Evaluasi](https://docs.claude.com/id/docs/test-and-evaluate/eval-tool.md): [Claude Console](https://console.anthropic.com/dashboard) menampilkan **alat Evaluasi** yang memungkinkan Anda menguji prompt Anda dalam berbagai skenario.
- [null](https://docs.claude.com/id/docs/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals.md)
- [Meningkatkan konsistensi output (mode JSON)](https://docs.claude.com/id/docs/test-and-evaluate/strengthen-guardrails/increase-consistency.md)
- [Menjaga Claude tetap dalam karakter dengan role prompting dan prefilling](https://docs.claude.com/id/docs/test-and-evaluate/strengthen-guardrails/keep-claude-in-character.md)
- [Mitigasi jailbreak dan injeksi prompt](https://docs.claude.com/id/docs/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks.md)
- [Mengurangi halusinasi](https://docs.claude.com/id/docs/test-and-evaluate/strengthen-guardrails/reduce-hallucinations.md)
- [Mengurangi latensi](https://docs.claude.com/id/docs/test-and-evaluate/strengthen-guardrails/reduce-latency.md)
- [Mengurangi kebocoran prompt](https://docs.claude.com/id/docs/test-and-evaluate/strengthen-guardrails/reduce-prompt-leak.md)
- [null](https://docs.claude.com/id/home.md)
- [Platform Pengembang Claude](https://docs.claude.com/id/release-notes/api.md): Ikuti pembaruan terbaru untuk API Claude dan Konsol Claude.
- [Aplikasi Claude](https://docs.claude.com/id/release-notes/claude-apps.md): Ikuti pembaruan terbaru di seluruh aplikasi Claude dari Anthropic.
- [Claude Code](https://docs.claude.com/id/release-notes/claude-code.md): Tetap terkini dengan fitur-fitur terbaru dan peningkatan pada Claude Code
- [Ikhtisar](https://docs.claude.com/id/release-notes/overview.md): Ikuti perkembangan terbaru di seluruh produk dan layanan Anthropic.
- [System Prompts](https://docs.claude.com/id/release-notes/system-prompts.md): Lihat pembaruan pada system prompts inti di [Claude.ai](https://www.claude.ai) dan aplikasi Claude [iOS](http://anthropic.com/ios) dan [Android](http://anthropic.com/android).
- [Kartu sistem Claude 3.7](https://docs.claude.com/id/resources/claude-3-7-system-card.md)
- [Kartu model Claude 3](https://docs.claude.com/id/resources/claude-3-model-card.md)
- [Kartu Sistem Claude 4](https://docs.claude.com/id/resources/claude-4-system-card.md): Kartu model Anthropic untuk Claude Opus dan Sonnet 4 dengan detail performa dan keamanan.
- [Anthropic Courses](https://docs.claude.com/id/resources/courses.md)
- [null](https://docs.claude.com/id/resources/overview.md)
- [Editor adaptif](https://docs.claude.com/id/resources/prompt-library/adaptive-editor.md): Tulis ulang teks mengikuti instruksi yang diberikan pengguna, seperti dengan nada, audiens, atau gaya yang berbeda.
- [Analis kode bandara](https://docs.claude.com/id/resources/prompt-library/airport-code-analyst.md): Temukan dan ekstrak kode bandara dari teks.
- [Antropolog alien](https://docs.claude.com/id/resources/prompt-library/alien-anthropologist.md): Analisis budaya dan adat istiadat manusia dari perspektif seorang antropolog alien.
- [Ahli alkimia aliterasi](https://docs.claude.com/id/resources/prompt-library/alliteration-alchemist.md): Menghasilkan frasa dan kalimat aliteratif untuk subjek yang diberikan.
- [Siaran Babel](https://docs.claude.com/id/resources/prompt-library/babels-broadcasts.md): Buat tweet pengumuman produk yang menarik dalam 10 bahasa yang paling banyak digunakan di dunia.
- [Pembangun merek](https://docs.claude.com/id/resources/prompt-library/brand-builder.md): Buat brief desain untuk identitas merek yang holistik.
- [Pelatih karir](https://docs.claude.com/id/resources/prompt-library/career-coach.md): Terlibat dalam percakapan bermain peran dengan pelatih karir AI.
- [Kutip sumber Anda](https://docs.claude.com/id/resources/prompt-library/cite-your-sources.md): Dapatkan jawaban atas pertanyaan tentang konten dokumen dengan kutipan relevan yang mendukung respons.
- [Penjelas kode](https://docs.claude.com/id/resources/prompt-library/code-clarifier.md): Sederhanakan dan jelaskan kode kompleks dalam bahasa yang mudah dipahami.
- [Konsultan kode](https://docs.claude.com/id/resources/prompt-library/code-consultant.md): Sarankan perbaikan untuk mengoptimalkan performa kode Python.
- [Peramal korporat](https://docs.claude.com/id/resources/prompt-library/corporate-clairvoyant.md): Ekstrak wawasan, identifikasi risiko, dan suling informasi kunci dari laporan korporat yang panjang menjadi satu memo.
- [Cosmic Keystrokes](https://docs.claude.com/id/resources/prompt-library/cosmic-keystrokes.md): Menghasilkan permainan mengetik kecepatan interaktif dalam satu file HTML, menampilkan gameplay side-scrolling dan styling Tailwind CSS.
- [Konverter CSV](https://docs.claude.com/id/resources/prompt-library/csv-converter.md): Mengonversi data dari berbagai format (JSON, XML, dll.) menjadi file CSV yang diformat dengan benar.
- [Pencipta kuliner](https://docs.claude.com/id/resources/prompt-library/culinary-creator.md): Sarankan ide resep berdasarkan bahan-bahan yang tersedia dan preferensi diet pengguna.
- [Pengorganisir data](https://docs.claude.com/id/resources/prompt-library/data-organizer.md): Ubah teks tidak terstruktur menjadi tabel JSON yang disesuaikan.
- [Dekoder arah](https://docs.claude.com/id/resources/prompt-library/direction-decoder.md): Mengubah bahasa alami menjadi petunjuk langkah demi langkah.
- [Penafsir mimpi](https://docs.claude.com/id/resources/prompt-library/dream-interpreter.md): Menawarkan interpretasi dan wawasan tentang simbolisme mimpi pengguna.
- [Estimator efisiensi](https://docs.claude.com/id/resources/prompt-library/efficiency-estimator.md): Hitung kompleksitas waktu dari fungsi dan algoritma.
- [Ekstraksi email](https://docs.claude.com/id/resources/prompt-library/email-extractor.md): Ekstrak alamat email dari dokumen ke dalam daftar berformat JSON.
- [Encoder emoji](https://docs.claude.com/id/resources/prompt-library/emoji-encoder.md): Mengubah teks biasa menjadi pesan emoji yang menyenangkan dan ekspresif.
- [Navigator dilema etis](https://docs.claude.com/id/resources/prompt-library/ethical-dilemma-navigator.md): Membantu pengguna memikirkan dilema etis yang kompleks dan memberikan perspektif yang berbeda.
- [Ahli formula Excel](https://docs.claude.com/id/resources/prompt-library/excel-formula-expert.md): Membuat formula Excel berdasarkan perhitungan atau manipulasi data yang dijelaskan pengguna.
- [Pembuat fungsi](https://docs.claude.com/id/resources/prompt-library/function-fabricator.md): Membuat fungsi Python berdasarkan spesifikasi yang detail.
- [Penasihat mode futuristik](https://docs.claude.com/id/resources/prompt-library/futuristic-fashion-advisor.md): Sarankan tren dan gaya mode avant-garde untuk preferensi spesifik pengguna.
- [Git gud](https://docs.claude.com/id/resources/prompt-library/git-gud.md): Menghasilkan perintah Git yang sesuai berdasarkan tindakan kontrol versi yang dijelaskan pengguna.
- [Pembuat skrip Google apps](https://docs.claude.com/id/resources/prompt-library/google-apps-scripter.md): Menghasilkan skrip Google Apps untuk menyelesaikan tugas berdasarkan kebutuhan pengguna.
- [Guru penilaian](https://docs.claude.com/id/resources/prompt-library/grading-guru.md): Membandingkan dan mengevaluasi kualitas teks tertulis berdasarkan kriteria dan standar yang ditentukan pengguna.
- [Grammar genie](https://docs.claude.com/id/resources/prompt-library/grammar-genie.md): Mengubah kalimat yang tidak benar secara tata bahasa menjadi bahasa Inggris yang tepat.
- [Hal si pembantu humoris](https://docs.claude.com/id/resources/prompt-library/hal-the-humorous-helper.md): Mengobrol dengan AI berpengetahuan luas yang memiliki sisi sarkastik.
- [Penjelas idiom](https://docs.claude.com/id/resources/prompt-library/idiom-illuminator.md): Menjelaskan makna dan asal-usul idiom dan peribahasa umum.
- [Pembuat pertanyaan wawancara](https://docs.claude.com/id/resources/prompt-library/interview-question-crafter.md): Menghasilkan pertanyaan untuk wawancara.
- [Legenda LaTeX](https://docs.claude.com/id/resources/prompt-library/latex-legend.md): Tulis dokumen LaTeX, menghasilkan kode untuk persamaan matematika, tabel, dan lainnya.
- [Perencana pelajaran](https://docs.claude.com/id/resources/prompt-library/lesson-planner.md): Buat rencana pelajaran mendalam tentang subjek apa pun.
- [Perpustakaan Prompt](https://docs.claude.com/id/resources/prompt-library/library.md)
- [Moderator utama](https://docs.claude.com/id/resources/prompt-library/master-moderator.md): Evaluasi input pengguna untuk konten yang berpotensi berbahaya atau ilegal.
- [Penulis rapat](https://docs.claude.com/id/resources/prompt-library/meeting-scribe.md): Meringkas rapat menjadi ringkasan yang ringkas termasuk topik diskusi, poin-poin penting, dan item tindakan.
- [Memo maestro](https://docs.claude.com/id/resources/prompt-library/memo-maestro.md): Menyusun memo perusahaan yang komprehensif berdasarkan poin-poin kunci.
- [Mentor mindfulness](https://docs.claude.com/id/resources/prompt-library/mindfulness-mentor.md): Membimbing pengguna melalui latihan mindfulness dan teknik untuk mengurangi stres.
- [Pewarna suasana hati](https://docs.claude.com/id/resources/prompt-library/mood-colorizer.md): Mengubah deskripsi teks suasana hati menjadi kode HEX yang sesuai.
- [Musa motivasional](https://docs.claude.com/id/resources/prompt-library/motivational-muse.md): Berikan pesan motivasi dan afirmasi yang dipersonalisasi berdasarkan masukan pengguna.
- [Pencipta neologisme](https://docs.claude.com/id/resources/prompt-library/neologism-creator.md): Menciptakan kata-kata baru dan memberikan definisinya berdasarkan konsep atau ide yang diberikan pengguna.
- [Pemikir perspektif](https://docs.claude.com/id/resources/prompt-library/perspectives-ponderer.md): Menimbang pro dan kontra dari topik yang diberikan pengguna.
- [Perenungan filosofis](https://docs.claude.com/id/resources/prompt-library/philosophical-musings.md): Terlibat dalam diskusi filosofis mendalam dan eksperimen pemikiran.
- [Pembersih PII](https://docs.claude.com/id/resources/prompt-library/pii-purifier.md): Secara otomatis mendeteksi dan menghapus informasi yang dapat mengidentifikasi pribadi (PII) dari dokumen teks.
- [Kekuatan super poliglot](https://docs.claude.com/id/resources/prompt-library/polyglot-superpowers.md): Terjemahkan teks dari bahasa apa pun ke bahasa apa pun.
- [Penyair portmanteau](https://docs.claude.com/id/resources/prompt-library/portmanteau-poet.md): Gabungkan dua kata bersama-sama untuk menciptakan portmanteau baru yang bermakna.
- [Ahli penamaan produk](https://docs.claude.com/id/resources/prompt-library/product-naming-pro.md): Buat nama produk yang menarik dari deskripsi dan kata kunci.
- [Penyempurna prosa](https://docs.claude.com/id/resources/prompt-library/prose-polisher.md): Perbaiki dan tingkatkan konten tertulis dengan teknik dan saran copyediting tingkat lanjut.
- [Pun-dit](https://docs.claude.com/id/resources/prompt-library/pun-dit.md): Menghasilkan permainan kata yang cerdas dan kreatif berdasarkan topik yang diberikan.
- [Pemburu bug Python](https://docs.claude.com/id/resources/prompt-library/python-bug-buster.md): Mendeteksi dan memperbaiki bug dalam kode Python.
- [Pengklasifikasi ulasan](https://docs.claude.com/id/resources/prompt-library/review-classifier.md): Kategorikan umpan balik ke dalam tag dan kategorisasi yang telah ditentukan sebelumnya.
- [Teka-teki untukku](https://docs.claude.com/id/resources/prompt-library/riddle-me-this.md): Buat teka-teki dan pandu pengguna menuju solusinya.
- [Simulator skenario sci-fi](https://docs.claude.com/id/resources/prompt-library/sci-fi-scenario-simulator.md): Diskusikan dengan pengguna berbagai skenario fiksi ilmiah dan tantangan serta pertimbangan yang terkait.
- [Penyederhanaan kelas dua](https://docs.claude.com/id/resources/prompt-library/second-grade-simplifier.md): Membuat teks kompleks mudah dipahami oleh pelajar muda.
- [Ahli perumpamaan](https://docs.claude.com/id/resources/prompt-library/simile-savant.md): Menghasilkan perumpamaan dari deskripsi dasar.
- [Bijak Socrates](https://docs.claude.com/id/resources/prompt-library/socratic-sage.md): Terlibat dalam percakapan gaya Socrates mengenai topik yang diberikan pengguna.
- [Penyihir spreadsheet](https://docs.claude.com/id/resources/prompt-library/spreadsheet-sorcerer.md): Menghasilkan spreadsheet CSV dengan berbagai jenis data.
- [Penyihir SQL](https://docs.claude.com/id/resources/prompt-library/sql-sorcerer.md): Mengubah bahasa sehari-hari menjadi query SQL.
- [Teman bercerita](https://docs.claude.com/id/resources/prompt-library/storytelling-sidekick.md): Berkolaborasi menciptakan cerita yang menarik dengan pengguna, menawarkan plot twist dan pengembangan karakter.
- [Konsultan perjalanan waktu](https://docs.claude.com/id/resources/prompt-library/time-travel-consultant.md): Membantu pengguna menavigasi skenario perjalanan waktu hipotetis dan implikasinya.
- [Tongue twister](https://docs.claude.com/id/resources/prompt-library/tongue-twister.md): Buat tongue twister yang menantang.
- [Generator trivia](https://docs.claude.com/id/resources/prompt-library/trivia-generator.md): Menghasilkan pertanyaan trivia tentang berbagai topik dan memberikan petunjuk ketika diperlukan.
- [Detektor nada tweet](https://docs.claude.com/id/resources/prompt-library/tweet-tone-detector.md): Mendeteksi nada dan sentimen di balik tweet.
- [Inovator kebugaran VR](https://docs.claude.com/id/resources/prompt-library/vr-fitness-innovator.md): Brainstorm ide-ide kreatif untuk permainan kebugaran virtual reality.
- [Wizard website](https://docs.claude.com/id/resources/prompt-library/website-wizard.md): Membuat website satu halaman berdasarkan spesifikasi pengguna.
- [Quickstarts](https://docs.claude.com/id/resources/quickstarts.md): Mulai dengan cepat menggunakan Claude dan Claude API
- [Status sistem](https://docs.claude.com/id/resources/status.md): Status sistem
- [Ottieni Chiave API](https://docs.claude.com/it/api/admin-api/apikeys/get-api-key.md)
- [Elenca Chiavi API](https://docs.claude.com/it/api/admin-api/apikeys/list-api-keys.md)
- [Aggiorna Chiavi API](https://docs.claude.com/it/api/admin-api/apikeys/update-api-key.md)
- [Ottieni Report di Utilizzo di Claude Code](https://docs.claude.com/it/api/admin-api/claude-code/get-claude-code-usage-report.md): Retrieve daily aggregated usage metrics for Claude Code users.
Enables organizations to analyze developer productivity and build custom dashboards.
- [Crea Invito](https://docs.claude.com/it/api/admin-api/invites/create-invite.md)
- [Elimina Invito](https://docs.claude.com/it/api/admin-api/invites/delete-invite.md)
- [Ottieni Invito](https://docs.claude.com/it/api/admin-api/invites/get-invite.md)
- [Elenca Inviti](https://docs.claude.com/it/api/admin-api/invites/list-invites.md)
- [Ottieni Informazioni Organizzazione](https://docs.claude.com/it/api/admin-api/organization/get-me.md)
- [Ottieni Rapporto Costi](https://docs.claude.com/it/api/admin-api/usage-cost/get-cost-report.md)
- [Ottieni Report di Utilizzo per l'API Messages](https://docs.claude.com/it/api/admin-api/usage-cost/get-messages-usage-report.md)
- [Ottieni Utente](https://docs.claude.com/it/api/admin-api/users/get-user.md)
- [Elenca Utenti](https://docs.claude.com/it/api/admin-api/users/list-users.md)
- [Rimuovi Utente](https://docs.claude.com/it/api/admin-api/users/remove-user.md)
- [Aggiorna Utente](https://docs.claude.com/it/api/admin-api/users/update-user.md)
- [Aggiungi Membro Workspace](https://docs.claude.com/it/api/admin-api/workspace_members/create-workspace-member.md)
- [Elimina Membro del Workspace](https://docs.claude.com/it/api/admin-api/workspace_members/delete-workspace-member.md)
- [Ottieni Membro del Workspace](https://docs.claude.com/it/api/admin-api/workspace_members/get-workspace-member.md)
- [Elenca Membri del Workspace](https://docs.claude.com/it/api/admin-api/workspace_members/list-workspace-members.md)
- [Aggiorna Membro Workspace](https://docs.claude.com/it/api/admin-api/workspace_members/update-workspace-member.md)
- [Archivia Workspace](https://docs.claude.com/it/api/admin-api/workspaces/archive-workspace.md)
- [Crea Workspace](https://docs.claude.com/it/api/admin-api/workspaces/create-workspace.md)
- [Ottieni Workspace](https://docs.claude.com/it/api/admin-api/workspaces/get-workspace.md)
- [Elenca Workspace](https://docs.claude.com/it/api/admin-api/workspaces/list-workspaces.md)
- [Aggiorna Workspace](https://docs.claude.com/it/api/admin-api/workspaces/update-workspace.md)
- [Panoramica dell'API Admin](https://docs.claude.com/it/api/administration-api.md)
- [Header beta](https://docs.claude.com/it/api/beta-headers.md): Documentazione per l'utilizzo degli header beta con l'API Claude
- [Annulla un Batch di Messaggi](https://docs.claude.com/it/api/canceling-message-batches.md): Batches may be canceled any time before processing ends. Once cancellation is initiated, the batch enters a `canceling` state, at which time the system may complete any in-progress, non-interruptible requests before finalizing cancellation.
The number of canceled requests is specified in `request_counts`. To determine which requests were canceled, check the individual results within the batch. Note that cancellation may not result in any canceled requests if they were non-interruptible.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [API di Analisi del Codice Claude](https://docs.claude.com/it/api/claude-code-analytics-api.md): Accedi programmaticamente alle analisi di utilizzo di Claude Code della tua organizzazione e alle metriche di produttività con l'API Admin di Analisi del Codice Claude.
- [API Amazon Bedrock](https://docs.claude.com/it/api/claude-on-amazon-bedrock.md): I modelli Claude di Anthropic sono ora generalmente disponibili tramite Amazon Bedrock.
- [API Vertex AI](https://docs.claude.com/it/api/claude-on-vertex-ai.md): I modelli Claude di Anthropic sono ora generalmente disponibili tramite [Vertex AI](https://cloud.google.com/vertex-ai).
- [SDK Client](https://docs.claude.com/it/api/client-sdks.md): Forniamo librerie client in diversi linguaggi popolari che rendono più facile lavorare con l'API Claude.
- [Crea un Batch di Messaggi](https://docs.claude.com/it/api/creating-message-batches.md): Send a batch of Message creation requests.
The Message Batches API can be used to process multiple Messages API requests at once. Once a Message Batch is created, it begins processing immediately. Batches can take up to 24 hours to complete.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Elimina un Batch di Messaggi](https://docs.claude.com/it/api/deleting-message-batches.md): Delete a Message Batch.
Message Batches can only be deleted once they've finished processing. If you'd like to delete an in-progress batch, you must first cancel it.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Errori](https://docs.claude.com/it/api/errors.md)
- [Scarica un File](https://docs.claude.com/it/api/files-content.md): Scarica i contenuti di un file generato da Claude
- [Crea un File](https://docs.claude.com/it/api/files-create.md): Carica un file
- [Elimina un File](https://docs.claude.com/it/api/files-delete.md): Rendi un file inaccessibile tramite l'API
- [Elenca File](https://docs.claude.com/it/api/files-list.md): Elenca i file all'interno di uno spazio di lavoro
- [Ottieni Metadati del File](https://docs.claude.com/it/api/files-metadata.md)
- [Ottenere aiuto](https://docs.claude.com/it/api/getting-help.md): Abbiamo cercato di fornire le risposte alle domande più comuni in questa documentazione. Tuttavia, se hai bisogno di ulteriore supporto tecnico per utilizzare Claude, l'API di Claude o qualsiasi dei nostri prodotti, puoi contattare il nostro team di supporto su [support.claude.com](https://support.claude.com).
- [Gestione dei motivi di arresto](https://docs.claude.com/it/api/handling-stop-reasons.md): Comprendi i valori stop_reason nell'API Messages di Claude e come gestire diversi tipi di risposta nelle tue applicazioni.
- [Indirizzi IP](https://docs.claude.com/it/api/ip-addresses.md): I servizi Anthropic utilizzano indirizzi IP fissi sia per le connessioni in entrata che in uscita. Puoi utilizzare questi indirizzi per configurare le regole del firewall per un accesso sicuro all'API Claude e alla Console. Questi indirizzi non cambieranno senza preavviso.
- [Elenca Batch di Messaggi](https://docs.claude.com/it/api/listing-message-batches.md): List all Message Batches within a Workspace. Most recently created batches are returned first.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Messaggi](https://docs.claude.com/it/api/messages.md): Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless multi-turn conversations.
Learn more about the Messages API in our [user guide](/en/docs/initial-setup)
- [Esempi di Message Batches](https://docs.claude.com/it/api/messages-batch-examples.md): Esempi di utilizzo per l'API Message Batches
- [Conta token messaggio](https://docs.claude.com/it/api/messages-count-tokens.md): Count the number of tokens in a Message.
The Token Count API can be used to count the number of tokens in a Message, including tools, images, and documents, without creating it.
Learn more about token counting in our [user guide](/en/docs/build-with-claude/token-counting)
- [Esempi di messaggi](https://docs.claude.com/it/api/messages-examples.md): Esempi di richiesta e risposta per l'API Messages
- [Migrazione da Text Completions](https://docs.claude.com/it/api/migrating-from-text-completions-to-messages.md): Migrazione da Text Completions a Messages
- [Ottieni un Modello](https://docs.claude.com/it/api/models.md): Get a specific model.
The Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.
- [Elenca Modelli](https://docs.claude.com/it/api/models-list.md): List available models.
The Models API response can be used to determine which models are available for use in the API. More recently released models are listed first.
- [Compatibilità SDK OpenAI](https://docs.claude.com/it/api/openai-sdk.md): Anthropic fornisce un livello di compatibilità che ti consente di utilizzare l'SDK OpenAI per testare l'API Claude. Con poche modifiche al codice, puoi valutare rapidamente le capacità del modello Anthropic.
- [Panoramica](https://docs.claude.com/it/api/overview.md): Panoramica dell'API Claude
- [Genera un prompt](https://docs.claude.com/it/api/prompt-tools-generate.md): Genera un prompt ben scritto
- [Migliorare un prompt](https://docs.claude.com/it/api/prompt-tools-improve.md): Creare un prompt nuovo e migliorato guidato dal feedback
- [Templatizzare un prompt](https://docs.claude.com/it/api/prompt-tools-templatize.md): Templatizzare un prompt identificando ed estraendo le variabili
- [Limiti di velocità](https://docs.claude.com/it/api/rate-limits.md): Per mitigare l'uso improprio e gestire la capacità sulla nostra API, abbiamo implementato limiti su quanto un'organizzazione può utilizzare l'API Claude.
- [Recupera Risultati Batch Messaggi](https://docs.claude.com/it/api/retrieving-message-batch-results.md): Streams the results of a Message Batch as a `.jsonl` file.
Each line in the file is a JSON object containing the result of a single request in the Message Batch. Results are not guaranteed to be in the same order as requests. Use the `custom_id` field to match results to requests.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Recupera un Batch di Messaggi](https://docs.claude.com/it/api/retrieving-message-batches.md): This endpoint is idempotent and can be used to poll for Message Batch completion. To access the results of a Message Batch, make a request to the `results_url` field in the response.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Livelli di servizio](https://docs.claude.com/it/api/service-tiers.md): Diversi livelli di servizio ti permettono di bilanciare disponibilità, prestazioni e costi prevedibili in base alle esigenze della tua applicazione.
- [Regioni supportate](https://docs.claude.com/it/api/supported-regions.md): Ecco i paesi, le regioni e i territori da cui attualmente possiamo supportare l'accesso:
- [API di Utilizzo e Costi](https://docs.claude.com/it/api/usage-cost-api.md): Accedi programmaticamente ai dati di utilizzo e costi API della tua organizzazione con l'API Admin di Utilizzo e Costi.
- [Versioni](https://docs.claude.com/it/api/versioning.md): Quando effettui richieste API, devi inviare un header di richiesta `anthropic-version`. Ad esempio, `anthropic-version: 2023-06-01`. Se stai utilizzando i nostri [SDK client](/it/api/client-sdks), questo viene gestito automaticamente per te.
- [Glossario](https://docs.claude.com/it/docs/about-claude/glossary.md): Questi concetti non sono unici ai modelli linguistici di Anthropic, ma presentiamo di seguito un breve riassunto dei termini chiave.
- [Deprecazioni dei modelli](https://docs.claude.com/it/docs/about-claude/model-deprecations.md)
- [Scegliere il modello giusto](https://docs.claude.com/it/docs/about-claude/models/choosing-a-model.md): Selezionare il modello Claude ottimale per la tua applicazione implica bilanciare tre considerazioni chiave: capacità, velocità e costo. Questa guida ti aiuta a prendere una decisione informata in base alle tue esigenze specifiche.
- [Migrazione a Claude 4](https://docs.claude.com/it/docs/about-claude/models/migrating-to-claude-4.md): Guida per migrare dai modelli Claude 3.7 ai modelli Claude 4 (Opus 4 e Sonnet 4).
- [Panoramica dei modelli](https://docs.claude.com/it/docs/about-claude/models/overview.md): Claude è una famiglia di modelli linguistici di grandi dimensioni all'avanguardia sviluppati da Anthropic. Questa guida introduce i nostri modelli e confronta le loro prestazioni con i modelli legacy.
- [Prezzi](https://docs.claude.com/it/docs/about-claude/pricing.md): Scopri la struttura dei prezzi di Anthropic per modelli e funzionalità
- [Moderazione dei contenuti](https://docs.claude.com/it/docs/about-claude/use-case-guides/content-moderation.md): La moderazione dei contenuti è un aspetto critico per mantenere un ambiente sicuro, rispettoso e produttivo nelle applicazioni digitali. In questa guida, discuteremo come Claude può essere utilizzato per moderare i contenuti all'interno della tua applicazione digitale.
- [Agente di supporto clienti](https://docs.claude.com/it/docs/about-claude/use-case-guides/customer-support-chat.md): Questa guida illustra come sfruttare le capacità conversazionali avanzate di Claude per gestire le richieste dei clienti in tempo reale, fornendo supporto 24/7, riducendo i tempi di attesa e gestendo volumi elevati di supporto con risposte accurate e interazioni positive.
- [Riassunto legale](https://docs.claude.com/it/docs/about-claude/use-case-guides/legal-summarization.md): Questa guida illustra come sfruttare le capacità avanzate di elaborazione del linguaggio naturale di Claude per riassumere efficacemente documenti legali, estraendo informazioni chiave e accelerando la ricerca legale. Con Claude, puoi semplificare la revisione di contratti, la preparazione di contenziosi e il lavoro normativo, risparmiando tempo e garantendo precisione nei tuoi processi legali.
- [Guide per i casi d'uso comuni](https://docs.claude.com/it/docs/about-claude/use-case-guides/overview.md)
- [Instradamento dei ticket](https://docs.claude.com/it/docs/about-claude/use-case-guides/ticket-routing.md): Questa guida illustra come sfruttare le capacità avanzate di comprensione del linguaggio naturale di Claude per classificare i ticket di supporto clienti su larga scala basandosi sull'intento del cliente, urgenza, prioritizzazione, profilo del cliente e altro ancora.
- [Componente aggiuntivo di Google Sheets](https://docs.claude.com/it/docs/agents-and-tools/claude-for-sheets.md): L'[estensione Claude for Sheets](https://workspace.google.com/marketplace/app/claude%5Ffor%5Fsheets/909417792257) integra Claude in Google Sheets, permettendoti di eseguire interazioni con Claude direttamente nelle celle.
- [Connettore MCP](https://docs.claude.com/it/docs/agents-and-tools/mcp-connector.md): Il connettore Model Context Protocol (MCP) di Claude ti consente di connetterti a server MCP remoti direttamente dall'API Messages senza un client MCP separato.
- [Server MCP remoti](https://docs.claude.com/it/docs/agents-and-tools/remote-mcp-servers.md): Server MCP remoti disponibili per sviluppatori
- [Strumento bash](https://docs.claude.com/it/docs/agents-and-tools/tool-use/bash-tool.md): Lo strumento bash consente a Claude di eseguire comandi shell in una sessione bash persistente, permettendo operazioni di sistema, esecuzione di script e automazione da riga di comando.
- [Strumento di esecuzione del codice](https://docs.claude.com/it/docs/agents-and-tools/tool-use/code-execution-tool.md)
- [Strumento di utilizzo del computer](https://docs.claude.com/it/docs/agents-and-tools/tool-use/computer-use-tool.md): Claude può interagire con ambienti informatici attraverso lo strumento di utilizzo del computer, che fornisce capacità di screenshot e controllo di mouse/tastiera per l'interazione autonoma del desktop.
- [Streaming granulare degli strumenti](https://docs.claude.com/it/docs/agents-and-tools/tool-use/fine-grained-tool-streaming.md): Scopri come utilizzare lo streaming granulare per i parametri degli strumenti per ridurre la latenza e ricevere parametri di grandi dimensioni più velocemente.
- [Come implementare l'uso degli strumenti](https://docs.claude.com/it/docs/agents-and-tools/tool-use/implement-tool-use.md)
- [Uso di strumenti con Claude](https://docs.claude.com/it/docs/agents-and-tools/tool-use/overview.md)
- [Strumento editor di testo](https://docs.claude.com/it/docs/agents-and-tools/tool-use/text-editor-tool.md): Claude può utilizzare uno strumento editor di testo definito da Anthropic per visualizzare e modificare file di testo, aiutandoti a debuggare, correggere e migliorare il tuo codice o altri documenti di testo.
- [Uso efficiente dei token per gli strumenti](https://docs.claude.com/it/docs/agents-and-tools/tool-use/token-efficient-tool-use.md): Scopri come utilizzare gli strumenti di Claude in modo efficiente dal punto di vista dei token per ridurre i costi e migliorare le prestazioni.
- [Strumento web fetch](https://docs.claude.com/it/docs/agents-and-tools/tool-use/web-fetch-tool.md): Lo strumento web fetch consente a Claude di recuperare contenuti completi da pagine web e documenti PDF specificati.
- [Strumento di ricerca web](https://docs.claude.com/it/docs/agents-and-tools/tool-use/web-search-tool.md): Lo strumento di ricerca web dà a Claude accesso diretto ai contenuti web in tempo reale, permettendogli di rispondere alle domande con informazioni aggiornate oltre il suo limite di conoscenza.
- [Elaborazione batch](https://docs.claude.com/it/docs/build-with-claude/batch-processing.md)
- [Citazioni](https://docs.claude.com/it/docs/build-with-claude/citations.md): Claude è in grado di fornire citazioni dettagliate quando risponde a domande sui documenti, aiutandoti a tracciare e verificare le fonti di informazione nelle risposte.
- [Finestre di contesto](https://docs.claude.com/it/docs/build-with-claude/context-windows.md): Comprendi come funzionano le finestre di contesto di Claude e come gestire i token in modo efficace.
- [Embeddings](https://docs.claude.com/it/docs/build-with-claude/embeddings.md): Gli embeddings di testo sono rappresentazioni numeriche del testo che consentono di misurare la similarità semantica. Questa guida introduce gli embeddings, le loro applicazioni e come utilizzare i modelli di embedding per compiti come ricerca, raccomandazioni e rilevamento di anomalie.
- [Costruire con il pensiero esteso](https://docs.claude.com/it/docs/build-with-claude/extended-thinking.md)
- [API Files](https://docs.claude.com/it/docs/build-with-claude/files.md)
- [Supporto multilingue](https://docs.claude.com/it/docs/build-with-claude/multilingual-support.md): Claude eccelle in compiti in più lingue, mantenendo prestazioni cross-linguistiche elevate rispetto all'inglese.
- [Panoramica delle funzionalità](https://docs.claude.com/it/docs/build-with-claude/overview.md): Esplora le funzionalità e capacità avanzate di Claude.
- [Supporto PDF](https://docs.claude.com/it/docs/build-with-claude/pdf-support.md): Elabora PDF con Claude. Estrai testo, analizza grafici e comprendi contenuti visivi dai tuoi documenti.
- [Caching dei prompt](https://docs.claude.com/it/docs/build-with-claude/prompt-caching.md)
- [Sii chiaro, diretto e dettagliato](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/be-clear-and-direct.md)
- [Lascia pensare Claude (chain of thought prompting) per migliorare le prestazioni](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/chain-of-thought.md)
- [Concatenare prompt complessi per prestazioni migliori](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/chain-prompts.md)
- [Migliori pratiche di prompt engineering per Claude 4](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/claude-4-best-practices.md): Tecniche specifiche di prompt engineering per i modelli Claude 4 per ottenere risultati ottimali nelle tue applicazioni.
- [Suggerimenti per il pensiero esteso](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/extended-thinking-tips.md)
- [Suggerimenti per il prompt con contesto lungo](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/long-context-tips.md)
- [Usa esempi (prompt multishot) per guidare il comportamento di Claude](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/multishot-prompting.md)
- [Panoramica dell'ingegneria dei prompt](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/overview.md)
- [Precompilare la risposta di Claude per un maggiore controllo dell'output](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/prefill-claudes-response.md)
- [Genera automaticamente modelli di prompt di prima bozza](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/prompt-generator.md)
- [Usa il nostro miglioratore di prompt per ottimizzare i tuoi prompt](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/prompt-improver.md): Usa il nostro miglioratore di prompt per ottimizzare i tuoi prompt
- [Usa template di prompt e variabili](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/prompt-templates-and-variables.md)
- [Assegnare un ruolo a Claude con un prompt di sistema](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/system-prompts.md)
- [Usa i tag XML per strutturare i tuoi prompt](https://docs.claude.com/it/docs/build-with-claude/prompt-engineering/use-xml-tags.md)
- [Risultati di ricerca](https://docs.claude.com/it/docs/build-with-claude/search-results.md): Abilita citazioni naturali per applicazioni RAG fornendo risultati di ricerca con attribuzione della fonte
- [Streaming dei Messaggi](https://docs.claude.com/it/docs/build-with-claude/streaming.md): Impara come utilizzare lo streaming per ricevere risposte incrementali dai messaggi di Claude utilizzando server-sent events.
- [Conteggio dei token](https://docs.claude.com/it/docs/build-with-claude/token-counting.md)
- [Visione](https://docs.claude.com/it/docs/build-with-claude/vision.md): Le famiglie di modelli Claude 3 e 4 sono dotate di nuove capacità di visione che permettono a Claude di comprendere e analizzare immagini, aprendo possibilità entusiasmanti per l'interazione multimodale.
- [Claude Code su Amazon Bedrock](https://docs.claude.com/it/docs/claude-code/amazon-bedrock.md): Scopri come configurare Claude Code tramite Amazon Bedrock, inclusi setup, configurazione IAM e risoluzione dei problemi.
- [Analisi](https://docs.claude.com/it/docs/claude-code/analytics.md): Visualizza informazioni dettagliate sull'utilizzo e metriche di produttività per il deployment di Claude Code della tua organizzazione.
- [Riferimento CLI](https://docs.claude.com/it/docs/claude-code/cli-reference.md): Riferimento completo per l'interfaccia a riga di comando di Claude Code, inclusi comandi e flag.
- [Flussi di lavoro comuni](https://docs.claude.com/it/docs/claude-code/common-workflows.md): Scopri i flussi di lavoro comuni con Claude Code.
- [Gestisci i costi in modo efficace](https://docs.claude.com/it/docs/claude-code/costs.md): Scopri come monitorare e ottimizzare l'utilizzo dei token e i costi quando usi Claude Code.
- [Utilizzo dei dati](https://docs.claude.com/it/docs/claude-code/data-usage.md): Scopri le politiche di utilizzo dei dati di Anthropic per Claude
- [Container di sviluppo](https://docs.claude.com/it/docs/claude-code/devcontainer.md): Scopri il container di sviluppo Claude Code per i team che necessitano di ambienti coerenti e sicuri.
- [Claude Code GitHub Actions](https://docs.claude.com/it/docs/claude-code/github-actions.md): Scopri come integrare Claude Code nel tuo flusso di lavoro di sviluppo con Claude Code GitHub Actions
- [Claude Code GitLab CI/CD](https://docs.claude.com/it/docs/claude-code/gitlab-ci-cd.md): Scopri come integrare Claude Code nel tuo flusso di lavoro di sviluppo con GitLab CI/CD
- [Claude Code su Google Vertex AI](https://docs.claude.com/it/docs/claude-code/google-vertex-ai.md): Scopri come configurare Claude Code tramite Google Vertex AI, inclusi setup, configurazione IAM e risoluzione dei problemi.
- [Riferimento hooks](https://docs.claude.com/it/docs/claude-code/hooks.md): Questa pagina fornisce documentazione di riferimento per l'implementazione di hooks in Claude Code.
- [Inizia con i hook di Claude Code](https://docs.claude.com/it/docs/claude-code/hooks-guide.md): Impara come personalizzare ed estendere il comportamento di Claude Code registrando comandi shell
- [Gestione delle Identità e degli Accessi](https://docs.claude.com/it/docs/claude-code/iam.md): Scopri come configurare l'autenticazione utente, l'autorizzazione e i controlli di accesso per Claude Code nella tua organizzazione.
- [Aggiungi Claude Code al tuo IDE](https://docs.claude.com/it/docs/claude-code/ide-integrations.md): Impara come aggiungere Claude Code al tuo IDE preferito
- [Modalità interattiva](https://docs.claude.com/it/docs/claude-code/interactive-mode.md): Riferimento completo per scorciatoie da tastiera, modalità di input e funzionalità interattive nelle sessioni di Claude Code.
- [Aspetti legali e conformità](https://docs.claude.com/it/docs/claude-code/legal-and-compliance.md): Accordi legali, certificazioni di conformità e informazioni sulla sicurezza per Claude Code.
- [Configurazione del gateway LLM](https://docs.claude.com/it/docs/claude-code/llm-gateway.md): Scopri come configurare Claude Code con soluzioni gateway LLM, inclusa la configurazione di LiteLLM, metodi di autenticazione e funzionalità aziendali come il monitoraggio dell'utilizzo e la gestione del budget.
- [Connetti Claude Code agli strumenti tramite MCP](https://docs.claude.com/it/docs/claude-code/mcp.md): Impara come connettere Claude Code ai tuoi strumenti con il Model Context Protocol.
- [Gestisci la memoria di Claude](https://docs.claude.com/it/docs/claude-code/memory.md): Impara come gestire la memoria di Claude Code attraverso le sessioni con diverse posizioni di memoria e migliori pratiche.
- [Configurazione del modello](https://docs.claude.com/it/docs/claude-code/model-config.md): Scopri la configurazione del modello Claude Code, inclusi gli alias del modello come `opusplan`
- [Monitoraggio](https://docs.claude.com/it/docs/claude-code/monitoring-usage.md): Scopri come abilitare e configurare OpenTelemetry per Claude Code.
- [Configurazione di rete aziendale](https://docs.claude.com/it/docs/claude-code/network-config.md): Configura Claude Code per ambienti aziendali con server proxy, Autorità di Certificazione (CA) personalizzate e autenticazione Transport Layer Security (mTLS) reciproca.
- [Stili di output](https://docs.claude.com/it/docs/claude-code/output-styles.md): Adatta Claude Code per usi oltre l'ingegneria del software
- [Panoramica di Claude Code](https://docs.claude.com/it/docs/claude-code/overview.md): Scopri Claude Code, lo strumento di codifica agentiva di Anthropic che vive nel tuo terminale e ti aiuta a trasformare le idee in codice più velocemente che mai.
- [Guida rapida](https://docs.claude.com/it/docs/claude-code/quickstart.md): Benvenuto in Claude Code!
- [Strumenti Personalizzati](https://docs.claude.com/it/docs/claude-code/sdk/custom-tools.md): Costruisci e integra strumenti personalizzati per estendere le funzionalità di Claude Code SDK
- [Modifica dei prompt di sistema](https://docs.claude.com/it/docs/claude-code/sdk/modifying-system-prompts.md): Impara come personalizzare il comportamento di Claude modificando i prompt di sistema utilizzando tre approcci - stili di output, appendSystemPrompt e customSystemPrompt.
- [Tracciamento dei Costi e dell'Utilizzo](https://docs.claude.com/it/docs/claude-code/sdk/sdk-cost-tracking.md): Comprendi e traccia l'utilizzo dei token per la fatturazione nell'SDK Claude Code
- [Modalità headless](https://docs.claude.com/it/docs/claude-code/sdk/sdk-headless.md): Esegui Claude Code programmaticamente senza interfaccia utente interattiva
- [MCP nell'SDK](https://docs.claude.com/it/docs/claude-code/sdk/sdk-mcp.md): Estendi Claude Code con strumenti personalizzati utilizzando i server Model Context Protocol
- [Panoramica](https://docs.claude.com/it/docs/claude-code/sdk/sdk-overview.md): Costruisci agenti AI personalizzati con l'SDK di Claude Code
- [Gestione dei Permessi](https://docs.claude.com/it/docs/claude-code/sdk/sdk-permissions.md): Controlla l'uso degli strumenti e i permessi nel Claude Code SDK
- [Riferimento SDK Python](https://docs.claude.com/it/docs/claude-code/sdk/sdk-python.md): Riferimento API completo per l'SDK Python di Claude Code, incluse tutte le funzioni, tipi e classi.
- [Gestione delle Sessioni](https://docs.claude.com/it/docs/claude-code/sdk/sdk-sessions.md): Comprendere come il Claude Code SDK gestisce le sessioni e la ripresa delle sessioni
- [Comandi Slash nell'SDK](https://docs.claude.com/it/docs/claude-code/sdk/sdk-slash-commands.md): Impara come utilizzare i comandi slash per controllare le sessioni di Claude Code attraverso l'SDK
- [Riferimento SDK TypeScript](https://docs.claude.com/it/docs/claude-code/sdk/sdk-typescript.md): Riferimento API completo per l'SDK TypeScript di Claude Code, incluse tutte le funzioni, tipi e interfacce.
- [Input in Streaming](https://docs.claude.com/it/docs/claude-code/sdk/streaming-vs-single-mode.md): Comprendere le due modalità di input per Claude Code SDK e quando utilizzare ciascuna
- [Subagenti nell'SDK](https://docs.claude.com/it/docs/claude-code/sdk/subagents.md): Lavorare con i subagenti nell'SDK di Claude Code
- [Liste Todo](https://docs.claude.com/it/docs/claude-code/sdk/todo-tracking.md): Traccia e visualizza i todo utilizzando il Claude Code SDK per una gestione organizzata delle attività
- [Sicurezza](https://docs.claude.com/it/docs/claude-code/security.md): Scopri le protezioni di sicurezza di Claude Code e le migliori pratiche per un utilizzo sicuro.
- [Impostazioni di Claude Code](https://docs.claude.com/it/docs/claude-code/settings.md): Configura Claude Code con impostazioni globali e a livello di progetto, e variabili d'ambiente.
- [Configurare Claude Code](https://docs.claude.com/it/docs/claude-code/setup.md): Installa, autentica e inizia a utilizzare Claude Code sulla tua macchina di sviluppo.
- [Comandi slash](https://docs.claude.com/it/docs/claude-code/slash-commands.md): Controlla il comportamento di Claude durante una sessione interattiva con i comandi slash.
- [Configurazione della riga di stato](https://docs.claude.com/it/docs/claude-code/statusline.md): Crea una riga di stato personalizzata per Claude Code per visualizzare informazioni contestuali
- [Subagenti](https://docs.claude.com/it/docs/claude-code/sub-agents.md): Crea e utilizza subagenti AI specializzati in Claude Code per flussi di lavoro specifici e una migliore gestione del contesto.
- [Ottimizza la configurazione del tuo terminale](https://docs.claude.com/it/docs/claude-code/terminal-config.md): Claude Code funziona meglio quando il tuo terminale è configurato correttamente. Segui queste linee guida per ottimizzare la tua esperienza.
- [Panoramica del deployment aziendale](https://docs.claude.com/it/docs/claude-code/third-party-integrations.md): Scopri come Claude Code può integrarsi con vari servizi di terze parti e infrastrutture per soddisfare i requisiti di deployment aziendale.
- [Risoluzione dei problemi](https://docs.claude.com/it/docs/claude-code/troubleshooting.md): Scopri soluzioni ai problemi comuni con l'installazione e l'uso di Claude Code.
- [Inizia con Claude](https://docs.claude.com/it/docs/get-started.md): Effettua la tua prima chiamata API a Claude e costruisci un semplice assistente di ricerca web
- [Introduzione a Claude](https://docs.claude.com/it/docs/intro.md): Claude è una piattaforma AI altamente performante, affidabile e intelligente costruita da Anthropic. Claude eccelle in compiti che coinvolgono linguaggio, ragionamento, analisi, programmazione e altro ancora.
- [Informativa sulla Privacy di Anthropic](https://docs.claude.com/it/docs/legal-center/privacy.md)
- [Sicurezza e conformità](https://docs.claude.com/it/docs/legal-center/security-compliance.md)
- [Costruire con Claude](https://docs.claude.com/it/docs/overview.md): Claude è una famiglia di [modelli AI altamente performanti e intelligenti](/it/docs/about-claude/models) costruiti da Anthropic. Sebbene Claude sia potente ed estensibile, è anche l'AI più affidabile e sicura disponibile. Segue protocolli critici, commette meno errori ed è resistente ai jailbreak—consentendo ai [clienti enterprise](https://claude.com/customers) di costruire le applicazioni AI più sicure su larga scala.
- [Definisci i tuoi criteri di successo](https://docs.claude.com/it/docs/test-and-evaluate/define-success.md)
- [Creare valutazioni empiriche solide](https://docs.claude.com/it/docs/test-and-evaluate/develop-tests.md): Sviluppa casi di test per misurare le prestazioni del tuo LLM contro i criteri di successo definiti.
- [Utilizzo dello Strumento di Valutazione](https://docs.claude.com/it/docs/test-and-evaluate/eval-tool.md): La [Claude Console](https://console.anthropic.com/dashboard) presenta uno **strumento di Valutazione** che ti permette di testare i tuoi prompt in vari scenari.
- [null](https://docs.claude.com/it/docs/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals.md)
- [Aumentare la coerenza dell'output (modalità JSON)](https://docs.claude.com/it/docs/test-and-evaluate/strengthen-guardrails/increase-consistency.md)
- [Mantenere Claude nel personaggio con il role prompting e il prefilling](https://docs.claude.com/it/docs/test-and-evaluate/strengthen-guardrails/keep-claude-in-character.md)
- [Mitigare i jailbreak e le iniezioni di prompt](https://docs.claude.com/it/docs/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks.md)
- [Ridurre le allucinazioni](https://docs.claude.com/it/docs/test-and-evaluate/strengthen-guardrails/reduce-hallucinations.md)
- [Riduzione della latenza](https://docs.claude.com/it/docs/test-and-evaluate/strengthen-guardrails/reduce-latency.md)
- [Ridurre la fuga di prompt](https://docs.claude.com/it/docs/test-and-evaluate/strengthen-guardrails/reduce-prompt-leak.md)
- [null](https://docs.claude.com/it/home.md)
- [Piattaforma per Sviluppatori Claude](https://docs.claude.com/it/release-notes/api.md): Segui gli aggiornamenti all'API Claude e alla Console Claude.
- [App Claude](https://docs.claude.com/it/release-notes/claude-apps.md): Segui gli aggiornamenti delle applicazioni Claude di Anthropic.
- [Claude Code](https://docs.claude.com/it/release-notes/claude-code.md): Rimani aggiornato con le ultime funzionalità e miglioramenti di Claude Code
- [Panoramica](https://docs.claude.com/it/release-notes/overview.md): Segui gli aggiornamenti sui prodotti e servizi di Anthropic.
- [Prompt di Sistema](https://docs.claude.com/it/release-notes/system-prompts.md): Vedi gli aggiornamenti ai prompt di sistema principali su [Claude.ai](https://www.claude.ai) e le app Claude per [iOS](http://anthropic.com/ios) e [Android](http://anthropic.com/android).
- [Scheda di sistema Claude 3.7](https://docs.claude.com/it/resources/claude-3-7-system-card.md)
- [Scheda modello Claude 3](https://docs.claude.com/it/resources/claude-3-model-card.md)
- [Scheda di Sistema Claude 4](https://docs.claude.com/it/resources/claude-4-system-card.md)
- [Corsi Anthropic](https://docs.claude.com/it/resources/courses.md)
- [null](https://docs.claude.com/it/resources/overview.md)
- [Editor adattivo](https://docs.claude.com/it/resources/prompt-library/adaptive-editor.md): Riscrivi il testo seguendo le istruzioni fornite dall'utente, come con un tono, pubblico o stile diverso.
- [Analizzatore di codici aeroportuali](https://docs.claude.com/it/resources/prompt-library/airport-code-analyst.md): Trova ed estrae i codici aeroportuali dal testo.
- [Antropologo alieno](https://docs.claude.com/it/resources/prompt-library/alien-anthropologist.md): Analizza la cultura e le usanze umane dalla prospettiva di un antropologo alieno.
- [Alchimista dell'allitterazione](https://docs.claude.com/it/resources/prompt-library/alliteration-alchemist.md): Genera frasi e periodi allitterativi per qualsiasi argomento dato.
- [Le trasmissioni di Babel](https://docs.claude.com/it/resources/prompt-library/babels-broadcasts.md): Crea tweet di annuncio prodotto coinvolgenti nelle 10 lingue più parlate al mondo.
- [Costruttore di brand](https://docs.claude.com/it/resources/prompt-library/brand-builder.md): Crea un brief di design per un'identità di brand olistica.
- [Coach di carriera](https://docs.claude.com/it/resources/prompt-library/career-coach.md): Partecipa a conversazioni di gioco di ruolo con un coach di carriera AI.
- [Cita le tue fonti](https://docs.claude.com/it/resources/prompt-library/cite-your-sources.md): Ottieni risposte a domande sul contenuto di un documento con citazioni rilevanti a supporto della risposta.
- [Chiarificatore di codice](https://docs.claude.com/it/resources/prompt-library/code-clarifier.md): Semplifica e spiega codice complesso in linguaggio semplice.
- [Consulente di codice](https://docs.claude.com/it/resources/prompt-library/code-consultant.md): Suggerisci miglioramenti per ottimizzare le prestazioni del codice Python.
- [Chiaroveggente aziendale](https://docs.claude.com/it/resources/prompt-library/corporate-clairvoyant.md): Estrai intuizioni, identifica rischi e distilla informazioni chiave da lunghi report aziendali in un singolo memo.
- [Cosmic Keystrokes](https://docs.claude.com/it/resources/prompt-library/cosmic-keystrokes.md): Genera un gioco interattivo di digitazione veloce in un singolo file HTML, con gameplay a scorrimento laterale e stile Tailwind CSS.
- [Convertitore CSV](https://docs.claude.com/it/resources/prompt-library/csv-converter.md): Converti dati da vari formati (JSON, XML, ecc.) in file CSV formattati correttamente.
- [Creatore culinario](https://docs.claude.com/it/resources/prompt-library/culinary-creator.md): Suggerisci idee per ricette basate sugli ingredienti disponibili dell'utente e sulle preferenze dietetiche.
- [Organizzatore di dati](https://docs.claude.com/it/resources/prompt-library/data-organizer.md): Trasforma testo non strutturato in tabelle JSON personalizzate.
- [Decodificatore di direzioni](https://docs.claude.com/it/resources/prompt-library/direction-decoder.md): Trasforma il linguaggio naturale in istruzioni passo dopo passo.
- [Interprete dei sogni](https://docs.claude.com/it/resources/prompt-library/dream-interpreter.md): Offri interpretazioni e intuizioni sul simbolismo dei sogni dell'utente.
- [Stimatore di efficienza](https://docs.claude.com/it/resources/prompt-library/efficiency-estimator.md): Calcola la complessità temporale di funzioni e algoritmi.
- [Estrattore di email](https://docs.claude.com/it/resources/prompt-library/email-extractor.md): Estrai indirizzi email da un documento in una lista formattata JSON.
- [Codificatore emoji](https://docs.claude.com/it/resources/prompt-library/emoji-encoder.md): Converti il testo semplice in messaggi emoji divertenti ed espressivi.
- [Navigatore di dilemmi etici](https://docs.claude.com/it/resources/prompt-library/ethical-dilemma-navigator.md): Aiuta l'utente a riflettere su dilemmi etici complessi e fornisce diverse prospettive.
- [Esperto di formule Excel](https://docs.claude.com/it/resources/prompt-library/excel-formula-expert.md): Crea formule Excel basate su calcoli o manipolazioni di dati descritti dall'utente.
- [Fabbricatore di funzioni](https://docs.claude.com/it/resources/prompt-library/function-fabricator.md): Crea funzioni Python basate su specifiche dettagliate.
- [Consulente di moda futuristica](https://docs.claude.com/it/resources/prompt-library/futuristic-fashion-advisor.md): Suggerisci tendenze e stili di moda d'avanguardia per le preferenze specifiche dell'utente.
- [Git gud](https://docs.claude.com/it/resources/prompt-library/git-gud.md): Genera comandi Git appropriati basati sulle azioni di controllo versione descritte dall'utente.
- [Scrittore di script per Google Apps](https://docs.claude.com/it/resources/prompt-library/google-apps-scripter.md): Genera script di Google Apps per completare attività basate sui requisiti dell'utente.
- [Guru della valutazione](https://docs.claude.com/it/resources/prompt-library/grading-guru.md): Confronta e valuta la qualità dei testi scritti basandosi su criteri e standard definiti dall'utente.
- [Genio della grammatica](https://docs.claude.com/it/resources/prompt-library/grammar-genie.md): Trasforma frasi grammaticalmente scorrette in inglese corretto.
- [Hal l'assistente umoristico](https://docs.claude.com/it/resources/prompt-library/hal-the-humorous-helper.md): Chatta con un'IA esperta che ha un lato sarcastico.
- [Illuminatore di modi di dire](https://docs.claude.com/it/resources/prompt-library/idiom-illuminator.md): Spiega il significato e l'origine di modi di dire e proverbi comuni.
- [Creatore di domande per colloqui](https://docs.claude.com/it/resources/prompt-library/interview-question-crafter.md): Genera domande per colloqui.
- [Leggenda LaTeX](https://docs.claude.com/it/resources/prompt-library/latex-legend.md): Scrivi documenti LaTeX, generando codice per equazioni matematiche, tabelle e altro.
- [Pianificatore di lezioni](https://docs.claude.com/it/resources/prompt-library/lesson-planner.md): Crea piani di lezione approfonditi su qualsiasi argomento.
- [Libreria di Prompt](https://docs.claude.com/it/resources/prompt-library/library.md)
- [Moderatore esperto](https://docs.claude.com/it/resources/prompt-library/master-moderator.md): Valuta gli input degli utenti per potenziali contenuti dannosi o illegali.
- [Segretario di riunioni](https://docs.claude.com/it/resources/prompt-library/meeting-scribe.md): Distilla le riunioni in riassunti concisi che includono argomenti di discussione, punti chiave e azioni da intraprendere.
- [Memo maestro](https://docs.claude.com/it/resources/prompt-library/memo-maestro.md): Componi memo aziendali completi basati su punti chiave.
- [Mentore di mindfulness](https://docs.claude.com/it/resources/prompt-library/mindfulness-mentor.md): Guida l'utente attraverso esercizi e tecniche di mindfulness per la riduzione dello stress.
- [Colorizzatore di umore](https://docs.claude.com/it/resources/prompt-library/mood-colorizer.md): Trasforma le descrizioni testuali degli umori in codici HEX corrispondenti.
- [Musa motivazionale](https://docs.claude.com/it/resources/prompt-library/motivational-muse.md): Fornisci messaggi motivazionali personalizzati e affermazioni positive in base all'input dell'utente.
- [Creatore di neologismi](https://docs.claude.com/it/resources/prompt-library/neologism-creator.md): Inventa nuove parole e fornisce le loro definizioni basate su concetti o idee fornite dall'utente.
- [Ponderatore di prospettive](https://docs.claude.com/it/resources/prompt-library/perspectives-ponderer.md): Valuta i pro e i contro di un argomento fornito dall'utente.
- [Riflessioni filosofiche](https://docs.claude.com/it/resources/prompt-library/philosophical-musings.md): Partecipa a discussioni filosofiche profonde ed esperimenti mentali.
- [Purificatore di PII](https://docs.claude.com/it/resources/prompt-library/pii-purifier.md): Rileva automaticamente e rimuove le informazioni di identificazione personale (PII) dai documenti di testo.
- [Superpoteri poliglotti](https://docs.claude.com/it/resources/prompt-library/polyglot-superpowers.md): Traduci testo da qualsiasi lingua in qualsiasi lingua.
- [Poeta di parole macedonia](https://docs.claude.com/it/resources/prompt-library/portmanteau-poet.md): Unisci due parole insieme per creare una nuova parola macedonia significativa.
- [Esperto di naming prodotti](https://docs.claude.com/it/resources/prompt-library/product-naming-pro.md): Crea nomi accattivanti per prodotti da descrizioni e parole chiave.
- [Levigatore di prosa](https://docs.claude.com/it/resources/prompt-library/prose-polisher.md): Rifinisci e migliora i contenuti scritti con tecniche avanzate di copyediting e suggerimenti.
- [Pun-dit](https://docs.claude.com/it/resources/prompt-library/pun-dit.md): Genera giochi di parole intelligenti e calembour basati su qualsiasi argomento dato.
- [Cacciatore di bug Python](https://docs.claude.com/it/resources/prompt-library/python-bug-buster.md): Rileva e corregge bug nel codice Python.
- [Classificatore di recensioni](https://docs.claude.com/it/resources/prompt-library/review-classifier.md): Categorizza i feedback in tag e categorizzazioni pre-specificati.
- [Indovinello per te](https://docs.claude.com/it/resources/prompt-library/riddle-me-this.md): Genera indovinelli e guida l'utente verso le soluzioni.
- [Simulatore di scenari fantascientifici](https://docs.claude.com/it/resources/prompt-library/sci-fi-scenario-simulator.md): Discuti con l'utente vari scenari di fantascienza e le sfide e considerazioni associate.
- [Semplificatore per seconda elementare](https://docs.claude.com/it/resources/prompt-library/second-grade-simplifier.md): Rendi i testi complessi facili da capire per i giovani studenti.
- [Esperto di similitudini](https://docs.claude.com/it/resources/prompt-library/simile-savant.md): Genera similitudini da descrizioni di base.
- [Saggio socratico](https://docs.claude.com/it/resources/prompt-library/socratic-sage.md): Partecipa a una conversazione in stile socratico su un argomento fornito dall'utente.
- [Mago dei fogli di calcolo](https://docs.claude.com/it/resources/prompt-library/spreadsheet-sorcerer.md): Genera fogli di calcolo CSV con vari tipi di dati.
- [Stregone SQL](https://docs.claude.com/it/resources/prompt-library/sql-sorcerer.md): Trasforma il linguaggio quotidiano in query SQL.
- [Compagno di narrazione](https://docs.claude.com/it/resources/prompt-library/storytelling-sidekick.md): Crea collaborativamente storie coinvolgenti con l'utente, offrendo colpi di scena e sviluppo dei personaggi.
- [Consulente di viaggi nel tempo](https://docs.claude.com/it/resources/prompt-library/time-travel-consultant.md): Aiuta l'utente a navigare scenari ipotetici di viaggi nel tempo e le loro implicazioni.
- [Scioglilingua](https://docs.claude.com/it/resources/prompt-library/tongue-twister.md): Crea scioglilingua impegnativi.
- [Generatore di trivia](https://docs.claude.com/it/resources/prompt-library/trivia-generator.md): Genera domande di trivia su una vasta gamma di argomenti e fornisce suggerimenti quando necessario.
- [Rilevatore di tono dei tweet](https://docs.claude.com/it/resources/prompt-library/tweet-tone-detector.md): Rileva il tono e il sentimento dietro i tweet.
- [Innovatore del fitness VR](https://docs.claude.com/it/resources/prompt-library/vr-fitness-innovator.md): Genera idee creative per giochi di fitness in realtà virtuale.
- [Procedura guidata per siti web](https://docs.claude.com/it/resources/prompt-library/website-wizard.md): Crea siti web di una sola pagina basati sulle specifiche dell'utente.
- [Avvii rapidi](https://docs.claude.com/it/resources/quickstarts.md): Inizia rapidamente con Claude e l'API Claude
- [Stato del sistema](https://docs.claude.com/it/resources/status.md)
- [APIキーを取得](https://docs.claude.com/ja/api/admin-api/apikeys/get-api-key.md)
- [APIキーの一覧表示](https://docs.claude.com/ja/api/admin-api/apikeys/list-api-keys.md)
- [APIキーの更新](https://docs.claude.com/ja/api/admin-api/apikeys/update-api-key.md)
- [Claude Code使用レポートを取得](https://docs.claude.com/ja/api/admin-api/claude-code/get-claude-code-usage-report.md): Retrieve daily aggregated usage metrics for Claude Code users.
Enables organizations to analyze developer productivity and build custom dashboards.
- [招待を作成](https://docs.claude.com/ja/api/admin-api/invites/create-invite.md): 組織への招待を作成します
- [招待を削除](https://docs.claude.com/ja/api/admin-api/invites/delete-invite.md)
- [招待を取得](https://docs.claude.com/ja/api/admin-api/invites/get-invite.md)
- [招待の一覧表示](https://docs.claude.com/ja/api/admin-api/invites/list-invites.md)
- [組織情報の取得](https://docs.claude.com/ja/api/admin-api/organization/get-me.md)
- [コストレポートを取得](https://docs.claude.com/ja/api/admin-api/usage-cost/get-cost-report.md)
- [Messages APIの使用状況レポートを取得](https://docs.claude.com/ja/api/admin-api/usage-cost/get-messages-usage-report.md)
- [ユーザーを取得](https://docs.claude.com/ja/api/admin-api/users/get-user.md)
- [ユーザー一覧](https://docs.claude.com/ja/api/admin-api/users/list-users.md)
- [ユーザーを削除](https://docs.claude.com/ja/api/admin-api/users/remove-user.md)
- [ユーザーを更新](https://docs.claude.com/ja/api/admin-api/users/update-user.md)
- [ワークスペースメンバーを追加](https://docs.claude.com/ja/api/admin-api/workspace_members/create-workspace-member.md): ワークスペースメンバーを追加
- [ワークスペースメンバーの削除](https://docs.claude.com/ja/api/admin-api/workspace_members/delete-workspace-member.md)
- [ワークスペースメンバーを取得](https://docs.claude.com/ja/api/admin-api/workspace_members/get-workspace-member.md)
- [ワークスペースメンバーの一覧表示](https://docs.claude.com/ja/api/admin-api/workspace_members/list-workspace-members.md)
- [ワークスペースメンバーの更新](https://docs.claude.com/ja/api/admin-api/workspace_members/update-workspace-member.md)
- [ワークスペースをアーカイブ](https://docs.claude.com/ja/api/admin-api/workspaces/archive-workspace.md)
- [ワークスペースを作成](https://docs.claude.com/ja/api/admin-api/workspaces/create-workspace.md)
- [ワークスペースを取得](https://docs.claude.com/ja/api/admin-api/workspaces/get-workspace.md)
- [ワークスペース一覧](https://docs.claude.com/ja/api/admin-api/workspaces/list-workspaces.md)
- [ワークスペースの更新](https://docs.claude.com/ja/api/admin-api/workspaces/update-workspace.md)
- [Admin API概要](https://docs.claude.com/ja/api/administration-api.md): Admin APIを使用して組織のリソース、メンバー、ワークスペース、APIキーをプログラムで管理する方法について学びます。
- [ベータヘッダー](https://docs.claude.com/ja/api/beta-headers.md): Claude APIでベータヘッダーを使用するためのドキュメント
- [メッセージバッチをキャンセル](https://docs.claude.com/ja/api/canceling-message-batches.md): Batches may be canceled any time before processing ends. Once cancellation is initiated, the batch enters a `canceling` state, at which time the system may complete any in-progress, non-interruptible requests before finalizing cancellation.
The number of canceled requests is specified in `request_counts`. To determine which requests were canceled, check the individual results within the batch. Note that cancellation may not result in any canceled requests if they were non-interruptible.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Claude Code Analytics API](https://docs.claude.com/ja/api/claude-code-analytics-api.md): Claude Code Analytics Admin APIを使用して、組織のClaude Codeの使用状況分析と生産性メトリクスにプログラム的にアクセスします。
- [Amazon Bedrock API](https://docs.claude.com/ja/api/claude-on-amazon-bedrock.md): AnthropicのClaudeモデルは、Amazon Bedrockを通じて一般提供されています。
- [Vertex AI API](https://docs.claude.com/ja/api/claude-on-vertex-ai.md): AnthropicのClaudeモデルは、[Vertex AI](https://cloud.google.com/vertex-ai)を通じて一般利用可能になりました。
- [クライアントSDK](https://docs.claude.com/ja/api/client-sdks.md): Claude APIをより簡単に使用できるよう、人気のある複数の言語でクライアントライブラリを提供しています。
- [メッセージバッチの作成](https://docs.claude.com/ja/api/creating-message-batches.md): Send a batch of Message creation requests.
The Message Batches API can be used to process multiple Messages API requests at once. Once a Message Batch is created, it begins processing immediately. Batches can take up to 24 hours to complete.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [メッセージバッチを削除](https://docs.claude.com/ja/api/deleting-message-batches.md): Delete a Message Batch.
Message Batches can only be deleted once they've finished processing. If you'd like to delete an in-progress batch, you must first cancel it.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [エラー](https://docs.claude.com/ja/api/errors.md)
- [ファイルをダウンロード](https://docs.claude.com/ja/api/files-content.md): Claude生成ファイルの内容をダウンロード
- [ファイルを作成](https://docs.claude.com/ja/api/files-create.md): ファイルをアップロード
- [ファイルを削除する](https://docs.claude.com/ja/api/files-delete.md): API経由でファイルにアクセスできないようにする
- [ファイル一覧](https://docs.claude.com/ja/api/files-list.md): ワークスペース内のファイルを一覧表示
- [ファイルメタデータの取得](https://docs.claude.com/ja/api/files-metadata.md)
- [ヘルプの取得](https://docs.claude.com/ja/api/getting-help.md): これらのドキュメントで最も一般的な質問への回答を提供するよう努めています。ただし、Claude、Claude API、または当社の製品の使用に関してさらなる技術サポートが必要な場合は、[support.claude.com](https://support.claude.com)で当社のサポートチームにお問い合わせください。
- [停止理由の処理](https://docs.claude.com/ja/api/handling-stop-reasons.md): Messages APIでClaude応答の停止理由を理解し、適切に処理する方法について説明します。
- [IPアドレス](https://docs.claude.com/ja/api/ip-addresses.md): Anthropicサービスは、インバウンドとアウトバウンドの両方の接続に固定IPアドレスを使用します。これらのアドレスを使用して、Claude APIとコンソールへの安全なアクセスのためのファイアウォールルールを設定できます。これらのアドレスは予告なしに変更されることはありません。
- [メッセージバッチ一覧](https://docs.claude.com/ja/api/listing-message-batches.md): List all Message Batches within a Workspace. Most recently created batches are returned first.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [メッセージ](https://docs.claude.com/ja/api/messages.md): Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless multi-turn conversations.
Learn more about the Messages API in our [user guide](/en/docs/initial-setup)
- [Message Batchesの例](https://docs.claude.com/ja/api/messages-batch-examples.md): Message Batches APIの使用例
- [メッセージトークンをカウント](https://docs.claude.com/ja/api/messages-count-tokens.md): Count the number of tokens in a Message.
The Token Count API can be used to count the number of tokens in a Message, including tools, images, and documents, without creating it.
Learn more about token counting in our [user guide](/en/docs/build-with-claude/token-counting)
- [Messagesの例](https://docs.claude.com/ja/api/messages-examples.md): Messages APIのリクエストとレスポンスの例
- [Text Completionsからの移行](https://docs.claude.com/ja/api/migrating-from-text-completions-to-messages.md): Text CompletionsからMessagesへの移行
- [モデルを取得](https://docs.claude.com/ja/api/models.md): Get a specific model.
The Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.
- [モデル一覧](https://docs.claude.com/ja/api/models-list.md): List available models.
The Models API response can be used to determine which models are available for use in the API. More recently released models are listed first.
- [OpenAI SDK互換性](https://docs.claude.com/ja/api/openai-sdk.md): AnthropicはOpenAI SDKを使用してClaude APIをテストできる互換性レイヤーを提供しています。わずかなコード変更で、Anthropicモデルの機能を迅速に評価できます。
- [概要](https://docs.claude.com/ja/api/overview.md): Claude APIの概要、認証、コンテンツタイプ、リクエストサイズ制限、レスポンスヘッダー、および使用例について説明します。
- [プロンプトを生成する](https://docs.claude.com/ja/api/prompt-tools-generate.md): よく書かれたプロンプトを生成する
- [プロンプトを改善する](https://docs.claude.com/ja/api/prompt-tools-improve.md): フィードバックに基づいて新しく改善されたプロンプトを作成する
- [プロンプトをテンプレート化する](https://docs.claude.com/ja/api/prompt-tools-templatize.md): 変数を特定して抽出することでプロンプトをテンプレート化する
- [レート制限](https://docs.claude.com/ja/api/rate-limits.md): APIの悪用を軽減し、容量を管理するため、組織がClaude APIを使用できる量に制限を設けています。
- [メッセージバッチ結果の取得](https://docs.claude.com/ja/api/retrieving-message-batch-results.md): Streams the results of a Message Batch as a `.jsonl` file.
Each line in the file is a JSON object containing the result of a single request in the Message Batch. Results are not guaranteed to be in the same order as requests. Use the `custom_id` field to match results to requests.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [メッセージバッチを取得](https://docs.claude.com/ja/api/retrieving-message-batches.md): This endpoint is idempotent and can be used to poll for Message Batch completion. To access the results of a Message Batch, make a request to the `results_url` field in the response.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [サービス階層](https://docs.claude.com/ja/api/service-tiers.md): 異なるサービス階層により、アプリケーションのニーズに基づいて可用性、パフォーマンス、予測可能なコストのバランスを取ることができます。
- [サポートされている地域](https://docs.claude.com/ja/api/supported-regions.md): 現在アクセスをサポートできる国、地域、および領土は以下の通りです：
- [使用量とコストAPI](https://docs.claude.com/ja/api/usage-cost-api.md): 使用量とコスト管理APIを使用して、組織のAPI使用量とコストデータにプログラム的にアクセスできます。
- [バージョン](https://docs.claude.com/ja/api/versioning.md): APIリクエストを行う際は、`anthropic-version`リクエストヘッダーを送信する必要があります。例：`anthropic-version: 2023-06-01`。[クライアントSDK](/ja/api/client-sdks)を使用している場合、これは自動的に処理されます。
- [用語集](https://docs.claude.com/ja/docs/about-claude/glossary.md): これらの概念はAnthropic社の言語モデルに固有のものではありませんが、以下に主要な用語の簡潔な要約を示します。
- [モデルの廃止予定](https://docs.claude.com/ja/docs/about-claude/model-deprecations.md)
- [適切なモデルの選択](https://docs.claude.com/ja/docs/about-claude/models/choosing-a-model.md): アプリケーションに最適なClaudeモデルを選択するには、機能、速度、コストという3つの重要な考慮事項のバランスを取る必要があります。このガイドは、特定の要件に基づいて情報に基づいた決定を下すのに役立ちます。
- [Claude 4への移行](https://docs.claude.com/ja/docs/about-claude/models/migrating-to-claude-4.md): Claude 3.7モデルからClaude 4モデル（Opus 4およびSonnet 4）への移行ガイダンス
- [モデル概要](https://docs.claude.com/ja/docs/about-claude/models/overview.md): Claudeは、Anthropicが開発した最先端の大規模言語モデルファミリーです。このガイドでは、私たちのモデルを紹介し、レガシーモデルとのパフォーマンスを比較します。
- [価格設定](https://docs.claude.com/ja/docs/about-claude/pricing.md): Anthropicのモデルと機能の価格構造について学ぶ
- [コンテンツモデレーション](https://docs.claude.com/ja/docs/about-claude/use-case-guides/content-moderation.md): コンテンツモデレーションは、デジタルアプリケーションにおいて安全で敬意に満ちた生産的な環境を維持するための重要な側面です。このガイドでは、デジタルアプリケーション内でコンテンツをモデレートするためにClaudeをどのように使用できるかについて説明します。
- [カスタマーサポートエージェント](https://docs.claude.com/ja/docs/about-claude/use-case-guides/customer-support-chat.md): このガイドでは、Claudeの高度な会話機能を活用してリアルタイムで顧客の問い合わせを処理し、24時間365日のサポートを提供し、待機時間を短縮し、正確な回答と前向きなやり取りで大量のサポート案件を管理する方法について説明します。
- [法的文書の要約](https://docs.claude.com/ja/docs/about-claude/use-case-guides/legal-summarization.md): このガイドでは、Claudeの高度な自然言語処理機能を活用して法的文書を効率的に要約し、重要な情報を抽出して法的調査を迅速化する方法について説明します。Claudeを使用することで、契約書のレビュー、訴訟準備、規制業務を合理化し、時間を節約しながら法的プロセスの正確性を確保できます。
- [一般的なユースケースのガイド](https://docs.claude.com/ja/docs/about-claude/use-case-guides/overview.md)
- [チケットルーティング](https://docs.claude.com/ja/docs/about-claude/use-case-guides/ticket-routing.md): このガイドでは、Claudeの高度な自然言語理解機能を活用して、顧客の意図、緊急度、優先度、顧客プロファイルなどに基づいて、カスタマーサポートチケットを大規模に分類する方法について説明します。
- [Google Sheetsアドオン](https://docs.claude.com/ja/docs/agents-and-tools/claude-for-sheets.md): [Claude for Sheets拡張機能](https://workspace.google.com/marketplace/app/claude%5Ffor%5Fsheets/909417792257)は、ClaudeをGoogle Sheetsに統合し、セル内でClaudeとの対話を直接実行できるようにします。
- [MCPコネクタ](https://docs.claude.com/ja/docs/agents-and-tools/mcp-connector.md): Claude の Model Context Protocol (MCP) コネクタ機能により、別途 MCP クライアントを使用せずに Messages API から直接リモート MCP サーバーに接続できます。
- [リモートMCPサーバー](https://docs.claude.com/ja/docs/agents-and-tools/remote-mcp-servers.md): 開発者がAnthropic MCP コネクタAPIを介して接続できるリモートMCPサーバーについて説明します。
- [Bashツール](https://docs.claude.com/ja/docs/agents-and-tools/tool-use/bash-tool.md): Claudeがシェルコマンドを実行し、システム操作、スクリプト実行、コマンドライン自動化を可能にするBashツールについて学びます。
- [コード実行ツール](https://docs.claude.com/ja/docs/agents-and-tools/tool-use/code-execution-tool.md)
- [コンピューター使用ツール](https://docs.claude.com/ja/docs/agents-and-tools/tool-use/computer-use-tool.md)
- [細粒度ツールストリーミング](https://docs.claude.com/ja/docs/agents-and-tools/tool-use/fine-grained-tool-streaming.md)
- [ツール使用の実装方法](https://docs.claude.com/ja/docs/agents-and-tools/tool-use/implement-tool-use.md): Claudeでツール使用を実装する方法について学びます。
- [Claudeでのツール使用](https://docs.claude.com/ja/docs/agents-and-tools/tool-use/overview.md)
- [テキストエディタツール](https://docs.claude.com/ja/docs/agents-and-tools/tool-use/text-editor-tool.md): Claudeがテキストファイルを表示・編集するためのAnthropic定義のテキストエディタツールの使用方法
- [トークン効率的なツール使用](https://docs.claude.com/ja/docs/agents-and-tools/tool-use/token-efficient-tool-use.md)
- [Web fetch ツール](https://docs.claude.com/ja/docs/agents-and-tools/tool-use/web-fetch-tool.md)
- [ウェブ検索ツール](https://docs.claude.com/ja/docs/agents-and-tools/tool-use/web-search-tool.md)
- [バッチ処理](https://docs.claude.com/ja/docs/build-with-claude/batch-processing.md)
- [引用](https://docs.claude.com/ja/docs/build-with-claude/citations.md): Claudeが文書に関する質問に答える際に詳細な引用を提供し、回答における情報源の追跡と検証を支援する機能について説明します。
- [コンテキストウィンドウ](https://docs.claude.com/ja/docs/build-with-claude/context-windows.md)
- [埋め込み](https://docs.claude.com/ja/docs/build-with-claude/embeddings.md): テキスト埋め込みは、意味的類似性を測定することを可能にするテキストの数値表現です。このガイドでは、埋め込み、その応用、および検索、推薦、異常検出などのタスクに埋め込みモデルを使用する方法を紹介します。
- [拡張思考による構築](https://docs.claude.com/ja/docs/build-with-claude/extended-thinking.md)
- [Files API](https://docs.claude.com/ja/docs/build-with-claude/files.md)
- [多言語サポート](https://docs.claude.com/ja/docs/build-with-claude/multilingual-support.md): Claudeは複数の言語にわたるタスクに優れており、英語と比較しても強力な言語間パフォーマンスを維持しています。
- [機能概要](https://docs.claude.com/ja/docs/build-with-claude/overview.md): Claudeの高度な機能と能力を探索します。
- [PDFサポート](https://docs.claude.com/ja/docs/build-with-claude/pdf-support.md): ClaudeでPDFを処理します。テキストの抽出、チャートの分析、ドキュメントからの視覚的コンテンツの理解を行います。
- [プロンプトキャッシュ](https://docs.claude.com/ja/docs/build-with-claude/prompt-caching.md): プロンプトキャッシュは、プロンプト内の特定のプレフィックスから再開することでAPIの使用を最適化する強力な機能です。
- [明確で直接的で詳細な指示を出す](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/be-clear-and-direct.md)
- [Claudeに考えさせる（思考連鎖プロンプト）でパフォーマンスを向上させる](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/chain-of-thought.md)
- [より強力なパフォーマンスのための複雑なプロンプトのチェーン化](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/chain-prompts.md)
- [Claude 4プロンプトエンジニアリングのベストプラクティス](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/claude-4-best-practices.md)
- [拡張思考のヒント](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/extended-thinking-tips.md)
- [長文コンテキストのプロンプト作成のヒント](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/long-context-tips.md)
- [例（マルチショットプロンプト）を使用してClaudeの動作を導く](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/multishot-prompting.md): 例を使用してClaudeの出力を正確に制御する方法を学びます。
- [プロンプトエンジニアリング概要](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/overview.md)
- [Claudeの応答を事前入力してより良い出力制御を実現する](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/prefill-claudes-response.md)
- [プロンプトテンプレートの初稿を自動生成](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/prompt-generator.md)
- [プロンプト改善ツールを使用してプロンプトを最適化する](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/prompt-improver.md)
- [プロンプトテンプレートと変数の使用](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/prompt-templates-and-variables.md): Claudeでプロンプトテンプレートと変数を使用する方法を学び、一貫性、効率性、テスト可能性を向上させます。
- [Claudeにシステムプロンプトで役割を与える](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/system-prompts.md)
- [XMLタグを使用してプロンプトを構造化する](https://docs.claude.com/ja/docs/build-with-claude/prompt-engineering/use-xml-tags.md)
- [検索結果](https://docs.claude.com/ja/docs/build-with-claude/search-results.md): ソース帰属を含む検索結果を提供することで、RAGアプリケーションの自然な引用を可能にします
- [ストリーミングメッセージ](https://docs.claude.com/ja/docs/build-with-claude/streaming.md): サーバー送信イベント（SSE）を使用してメッセージレスポンスを段階的にストリーミングする方法について説明します。
- [トークンカウント](https://docs.claude.com/ja/docs/build-with-claude/token-counting.md)
- [ビジョン](https://docs.claude.com/ja/docs/build-with-claude/vision.md): Claude 3および4ファミリーのモデルには、Claudeが画像を理解し分析できる新しいビジョン機能が搭載されており、マルチモーダルインタラクションの刺激的な可能性を開きます。
- [Amazon BedrockでのClaude Code](https://docs.claude.com/ja/docs/claude-code/amazon-bedrock.md): Amazon Bedrockを通じたClaude Codeの設定について、セットアップ、IAM設定、トラブルシューティングを含めて学習します。
- [アナリティクス](https://docs.claude.com/ja/docs/claude-code/analytics.md): 組織のClaude Codeデプロイメントの詳細な使用状況インサイトと生産性メトリクスを表示します。
- [CLIリファレンス](https://docs.claude.com/ja/docs/claude-code/cli-reference.md): Claude Codeコマンドラインインターフェースの完全なリファレンス（コマンドとフラグを含む）。
- [一般的なワークフロー](https://docs.claude.com/ja/docs/claude-code/common-workflows.md): Claude Codeを使った一般的なワークフローについて学びます。
- [コストを効果的に管理する](https://docs.claude.com/ja/docs/claude-code/costs.md): Claude Codeを使用する際のトークン使用量とコストを追跡し最適化する方法を学びます。
- [データ使用](https://docs.claude.com/ja/docs/claude-code/data-usage.md): ClaudeのAnthropicのデータ使用ポリシーについて学ぶ
- [開発コンテナ](https://docs.claude.com/ja/docs/claude-code/devcontainer.md): 一貫性のある安全な環境が必要なチーム向けのClaude Code開発コンテナについて学びます。
- [Claude Code GitHub Actions](https://docs.claude.com/ja/docs/claude-code/github-actions.md): Claude Code GitHub Actionsを使用して、Claude Codeを開発ワークフローに統合する方法について学びます
- [Claude Code GitLab CI/CD](https://docs.claude.com/ja/docs/claude-code/gitlab-ci-cd.md): GitLab CI/CDを使用してClaude Codeを開発ワークフローに統合する方法について学習します
- [Google Vertex AI上のClaude Code](https://docs.claude.com/ja/docs/claude-code/google-vertex-ai.md): Google Vertex AIを通じたClaude Codeの設定について、セットアップ、IAM設定、トラブルシューティングを含めて学習します。
- [フック リファレンス](https://docs.claude.com/ja/docs/claude-code/hooks.md): このページでは、Claude Code でフックを実装するためのリファレンス ドキュメントを提供します。
- [Claude Code フックを始める](https://docs.claude.com/ja/docs/claude-code/hooks-guide.md): シェルコマンドを登録してClaude Codeの動作をカスタマイズおよび拡張する方法を学ぶ
- [アイデンティティとアクセス管理](https://docs.claude.com/ja/docs/claude-code/iam.md): 組織内でのClaude Codeのユーザー認証、認可、およびアクセス制御の設定方法を学びます。
- [IDEにClaude Codeを追加する](https://docs.claude.com/ja/docs/claude-code/ide-integrations.md): お気に入りのIDEにClaude Codeを追加する方法を学ぶ
- [インタラクティブモード](https://docs.claude.com/ja/docs/claude-code/interactive-mode.md): Claude Codeセッションでのキーボードショートカット、入力モード、インタラクティブ機能の完全なリファレンス。
- [法的事項とコンプライアンス](https://docs.claude.com/ja/docs/claude-code/legal-and-compliance.md): Claude Codeの法的合意、コンプライアンス認証、およびセキュリティ情報。
- [LLMゲートウェイ設定](https://docs.claude.com/ja/docs/claude-code/llm-gateway.md): LiteLLMセットアップ、認証方法、使用量追跡や予算管理などのエンタープライズ機能を含む、LLMゲートウェイソリューションでClaude Codeを設定する方法を学びます。
- [MCPを介してClaude Codeをツールに接続する](https://docs.claude.com/ja/docs/claude-code/mcp.md): Model Context ProtocolでClaude Codeをあなたのツールに接続する方法を学びます。
- [Claudeのメモリを管理する](https://docs.claude.com/ja/docs/claude-code/memory.md): 異なるメモリ場所とベストプラクティスを使用して、セッション間でClaude Codeのメモリを管理する方法を学びます。
- [モデル設定](https://docs.claude.com/ja/docs/claude-code/model-config.md): Claude Codeのモデル設定について学びます。`opusplan`などのモデルエイリアスを含みます。
- [モニタリング](https://docs.claude.com/ja/docs/claude-code/monitoring-usage.md): Claude CodeでOpenTelemetryを有効化し設定する方法を学びます。
- [エンタープライズネットワーク設定](https://docs.claude.com/ja/docs/claude-code/network-config.md): プロキシサーバー、カスタム認証局（CA）、および相互トランスポート層セキュリティ（mTLS）認証を使用したエンタープライズ環境向けのClaude Codeの設定。
- [出力スタイル](https://docs.claude.com/ja/docs/claude-code/output-styles.md): Claude Codeをソフトウェアエンジニアリング以外の用途に適応させる
- [Claude Code概要](https://docs.claude.com/ja/docs/claude-code/overview.md): ターミナルで動作し、アイデアをこれまで以上に高速でコードに変換するのを支援するAnthropicのエージェンティックコーディングツール、Claude Codeについて学びます。
- [クイックスタート](https://docs.claude.com/ja/docs/claude-code/quickstart.md): Claude Codeへようこそ！
- [カスタムツール](https://docs.claude.com/ja/docs/claude-code/sdk/custom-tools.md): Claude Code SDKの機能を拡張するためのカスタムツールを構築・統合する
- [システムプロンプトの変更](https://docs.claude.com/ja/docs/claude-code/sdk/modifying-system-prompts.md): 出力スタイル、appendSystemPrompt、customSystemPromptの3つのアプローチを使用してシステムプロンプトを変更し、Claudeの動作をカスタマイズする方法を学びます。
- [コストと使用量の追跡](https://docs.claude.com/ja/docs/claude-code/sdk/sdk-cost-tracking.md): Claude Code SDKでの請求のためのトークン使用量の理解と追跡
- [ヘッドレスモード](https://docs.claude.com/ja/docs/claude-code/sdk/sdk-headless.md): インタラクティブUIなしでClaude Codeをプログラム的に実行する
- [SDKにおけるMCP](https://docs.claude.com/ja/docs/claude-code/sdk/sdk-mcp.md): Model Context Protocolサーバーを使用してClaude Codeをカスタムツールで拡張する
- [概要](https://docs.claude.com/ja/docs/claude-code/sdk/sdk-overview.md): Claude Code SDKでカスタムAIエージェントを構築する
- [権限の処理](https://docs.claude.com/ja/docs/claude-code/sdk/sdk-permissions.md): Claude Code SDKでのツール使用と権限の制御
- [Python SDK リファレンス](https://docs.claude.com/ja/docs/claude-code/sdk/sdk-python.md): Claude Code Python SDK の完全な API リファレンス。すべての関数、型、クラスを含みます。
- [セッション管理](https://docs.claude.com/ja/docs/claude-code/sdk/sdk-sessions.md): Claude Code SDKがセッションとセッション再開をどのように処理するかを理解する
- [SDKでのスラッシュコマンド](https://docs.claude.com/ja/docs/claude-code/sdk/sdk-slash-commands.md): SDKを通じてClaude Codeセッションを制御するためのスラッシュコマンドの使用方法を学ぶ
- [TypeScript SDK リファレンス](https://docs.claude.com/ja/docs/claude-code/sdk/sdk-typescript.md): Claude Code TypeScript SDK の完全な API リファレンス。すべての関数、型、インターフェースを含みます。
- [ストリーミング入力](https://docs.claude.com/ja/docs/claude-code/sdk/streaming-vs-single-mode.md): Claude Code SDKの2つの入力モードとそれぞれをいつ使用するかの理解
- [SDKにおけるサブエージェント](https://docs.claude.com/ja/docs/claude-code/sdk/subagents.md): Claude Code SDKでサブエージェントを使用する方法
- [Todo リスト](https://docs.claude.com/ja/docs/claude-code/sdk/todo-tracking.md): 整理されたタスク管理のためにClaude Code SDKを使用してtodoを追跡・表示する
- [セキュリティ](https://docs.claude.com/ja/docs/claude-code/security.md): Claude Codeのセキュリティ保護機能と安全な使用のためのベストプラクティスについて学びます。
- [Claude Code設定](https://docs.claude.com/ja/docs/claude-code/settings.md): グローバルおよびプロジェクトレベルの設定、環境変数でClaude Codeを設定します。
- [Claude Codeのセットアップ](https://docs.claude.com/ja/docs/claude-code/setup.md): 開発マシンでClaude Codeをインストール、認証、使用開始する方法。
- [スラッシュコマンド](https://docs.claude.com/ja/docs/claude-code/slash-commands.md): インタラクティブセッション中にスラッシュコマンドでClaudeの動作を制御します。
- [ステータスライン設定](https://docs.claude.com/ja/docs/claude-code/statusline.md): Claude Codeでコンテキスト情報を表示するカスタムステータスラインを作成する
- [サブエージェント](https://docs.claude.com/ja/docs/claude-code/sub-agents.md): Claude Codeでタスク固有のワークフローと改善されたコンテキスト管理のための専門的なAIサブエージェントを作成・使用する。
- [ターミナル設定を最適化する](https://docs.claude.com/ja/docs/claude-code/terminal-config.md): Claude Codeは、ターミナルが適切に設定されている場合に最高のパフォーマンスを発揮します。これらのガイドラインに従って、体験を最適化してください。
- [エンタープライズデプロイメント概要](https://docs.claude.com/ja/docs/claude-code/third-party-integrations.md): Claude Codeがさまざまなサードパーティサービスやインフラストラクチャと統合して、エンタープライズデプロイメント要件を満たす方法を学びます。
- [トラブルシューティング](https://docs.claude.com/ja/docs/claude-code/troubleshooting.md): Claude Codeのインストールと使用に関する一般的な問題の解決策を発見してください。
- [Claudeを始める](https://docs.claude.com/ja/docs/get-started.md): Claudeへの最初のAPI呼び出しを行い、シンプルなウェブ検索アシスタントを構築する
- [Claudeの紹介](https://docs.claude.com/ja/docs/intro.md): Claudeは、Anthropicが構築した高性能で信頼性が高く、インテリジェントなAIプラットフォームです。Claudeは言語、推論、分析、コーディングなどのタスクに優れています。
- [Anthropic プライバシーポリシー](https://docs.claude.com/ja/docs/legal-center/privacy.md)
- [セキュリティとコンプライアンス](https://docs.claude.com/ja/docs/legal-center/security-compliance.md)
- [Claudeでの構築](https://docs.claude.com/ja/docs/overview.md): Claudeは、Anthropicが構築した[高性能で知的なAIモデル](/ja/docs/about-claude/models)のファミリーです。Claudeは強力で拡張可能でありながら、利用可能な最も信頼できる確実なAIでもあります。重要なプロトコルに従い、ミスが少なく、ジェイルブレイクに耐性があるため、[エンタープライズ顧客](https://claude.com/customers)が最も安全なAI搭載アプリケーションを大規模に構築できます。
- [成功基準を定義する](https://docs.claude.com/ja/docs/test-and-evaluate/define-success.md): LLMベースのアプリケーションを構築する際の明確な成功基準の設定方法について学びます。
- [強力な実証的評価を作成する](https://docs.claude.com/ja/docs/test-and-evaluate/develop-tests.md): 成功基準を定義した後、次のステップはそれらの基準に対してLLMのパフォーマンスを測定する評価を設計することです。これはプロンプトエンジニアリングサイクルの重要な部分です。
- [評価ツールの使用](https://docs.claude.com/ja/docs/test-and-evaluate/eval-tool.md): [Claude Console](https://console.anthropic.com/dashboard)には、様々なシナリオでプロンプトをテストできる**評価ツール**が搭載されています。
- [null](https://docs.claude.com/ja/docs/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals.md)
- [出力の一貫性を高める（JSONモード）](https://docs.claude.com/ja/docs/test-and-evaluate/strengthen-guardrails/increase-consistency.md)
- [ロールプロンプトとプリフィルでClaudeのキャラクターを維持する](https://docs.claude.com/ja/docs/test-and-evaluate/strengthen-guardrails/keep-claude-in-character.md)
- [ジェイルブレイクとプロンプトインジェクションを軽減する](https://docs.claude.com/ja/docs/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks.md)
- [幻覚を減らす](https://docs.claude.com/ja/docs/test-and-evaluate/strengthen-guardrails/reduce-hallucinations.md)
- [レイテンシーの削減](https://docs.claude.com/ja/docs/test-and-evaluate/strengthen-guardrails/reduce-latency.md)
- [プロンプトリークの削減](https://docs.claude.com/ja/docs/test-and-evaluate/strengthen-guardrails/reduce-prompt-leak.md)
- [null](https://docs.claude.com/ja/home.md)
- [Claude Developer Platform](https://docs.claude.com/ja/release-notes/api.md): Claude APIとClaude Consoleのアップデートをフォローしてください。
- [Claude Apps](https://docs.claude.com/ja/release-notes/claude-apps.md): AnthropicのClaudeアプリケーション全体のアップデートをフォローしてください。
- [Claude Code](https://docs.claude.com/ja/release-notes/claude-code.md): Claude Codeの最新機能と改善点について最新情報を入手する
- [概要](https://docs.claude.com/ja/release-notes/overview.md): Anthropicの製品とサービス全体のアップデートをフォローしてください。
- [システムプロンプト](https://docs.claude.com/ja/release-notes/system-prompts.md): [Claude.ai](https://www.claude.ai)およびClaude [iOS](http://anthropic.com/ios)と[Android](http://anthropic.com/android)アプリのコアシステムプロンプトの更新をご覧ください。
- [Claude 3.7 システムカード](https://docs.claude.com/ja/resources/claude-3-7-system-card.md)
- [Claude 3モデルカード](https://docs.claude.com/ja/resources/claude-3-model-card.md)
- [Claude 4 システムカード](https://docs.claude.com/ja/resources/claude-4-system-card.md): Claude Opus と Sonnet 4 のパフォーマンスと安全性の詳細を含む Anthropic のモデルカード。
- [Anthropic コース](https://docs.claude.com/ja/resources/courses.md)
- [null](https://docs.claude.com/ja/resources/overview.md)
- [適応型エディター](https://docs.claude.com/ja/resources/prompt-library/adaptive-editor.md): 異なるトーン、対象読者、スタイルなど、ユーザーが指定した指示に従ってテキストを書き直します。
- [空港コード分析器](https://docs.claude.com/ja/resources/prompt-library/airport-code-analyst.md): テキストから空港コードを見つけて抽出します。
- [エイリアン人類学者](https://docs.claude.com/ja/resources/prompt-library/alien-anthropologist.md): エイリアン人類学者の視点から人間の文化と習慣を分析します。
- [頭韻錬金術師](https://docs.claude.com/ja/resources/prompt-library/alliteration-alchemist.md): 任意の主題に対して頭韻を踏んだフレーズや文章を生成します。
- [Babelの放送](https://docs.claude.com/ja/resources/prompt-library/babels-broadcasts.md): 世界で最も話されている10の言語で魅力的な製品発表ツイートを作成します。
- [ブランドビルダー](https://docs.claude.com/ja/resources/prompt-library/brand-builder.md): 包括的なブランドアイデンティティのためのデザインブリーフを作成します。
- [キャリアコーチ](https://docs.claude.com/ja/resources/prompt-library/career-coach.md): AIキャリアコーチとのロールプレイ会話を行います。
- [情報源を引用する](https://docs.claude.com/ja/resources/prompt-library/cite-your-sources.md): 文書の内容に関する質問に対して、回答を支持する関連する引用とともに答えを得る。
- [コード解説器](https://docs.claude.com/ja/resources/prompt-library/code-clarifier.md): 複雑なコードを平易な言葉で簡潔に説明します。
- [コードコンサルタント](https://docs.claude.com/ja/resources/prompt-library/code-consultant.md): Pythonコードのパフォーマンスを最適化するための改善案を提案します。
- [企業の千里眼](https://docs.claude.com/ja/resources/prompt-library/corporate-clairvoyant.md): 長い企業レポートから洞察を抽出し、リスクを特定し、重要な情報を1つのメモに要約します。
- [コズミック・キーストローク](https://docs.claude.com/ja/resources/prompt-library/cosmic-keystrokes.md): サイドスクロールゲームプレイとTailwind CSSスタイリングを特徴とする、インタラクティブなスピードタイピングゲームを単一のHTMLファイルで生成します。
- [CSVコンバーター](https://docs.claude.com/ja/resources/prompt-library/csv-converter.md): 様々な形式（JSON、XMLなど）のデータを適切にフォーマットされたCSVファイルに変換します。
- [料理クリエイター](https://docs.claude.com/ja/resources/prompt-library/culinary-creator.md): ユーザーの利用可能な食材と食事の好みに基づいてレシピのアイデアを提案します。
- [データ整理ツール](https://docs.claude.com/ja/resources/prompt-library/data-organizer.md): 非構造化テキストをカスタムJSONテーブルに変換します。
- [方向デコーダー](https://docs.claude.com/ja/resources/prompt-library/direction-decoder.md): 自然言語を段階的な手順に変換します。
- [夢の解釈者](https://docs.claude.com/ja/resources/prompt-library/dream-interpreter.md): ユーザーの夢の象徴性について解釈と洞察を提供します。
- [効率性推定器](https://docs.claude.com/ja/resources/prompt-library/efficiency-estimator.md): 関数とアルゴリズムの時間計算量を計算します。
- [メール抽出器](https://docs.claude.com/ja/resources/prompt-library/email-extractor.md): 文書からメールアドレスを抽出してJSON形式のリストにします。
- [絵文字エンコーダー](https://docs.claude.com/ja/resources/prompt-library/emoji-encoder.md): プレーンテキストを楽しく表現豊かな絵文字メッセージに変換します。
- [倫理的ジレンマナビゲーター](https://docs.claude.com/ja/resources/prompt-library/ethical-dilemma-navigator.md): 複雑な倫理的ジレンマを考え抜き、さまざまな視点を提供することでユーザーを支援します。
- [Excel数式エキスパート](https://docs.claude.com/ja/resources/prompt-library/excel-formula-expert.md): ユーザーが説明する計算やデータ操作に基づいてExcel数式を作成します。
- [関数ファブリケーター](https://docs.claude.com/ja/resources/prompt-library/function-fabricator.md): 詳細な仕様に基づいてPython関数を作成します。
- [未来的ファッションアドバイザー](https://docs.claude.com/ja/resources/prompt-library/futuristic-fashion-advisor.md): ユーザーの特定の好みに合わせたアバンギャルドなファッショントレンドとスタイルを提案します。
- [Git gud](https://docs.claude.com/ja/resources/prompt-library/git-gud.md): ユーザーが説明するバージョン管理アクションに基づいて適切なGitコマンドを生成します。
- [Google Apps Scripter](https://docs.claude.com/ja/resources/prompt-library/google-apps-scripter.md): ユーザーの要件に基づいてタスクを完了するためのGoogle Apps Scriptを生成します。
- [採点の達人](https://docs.claude.com/ja/resources/prompt-library/grading-guru.md): ユーザー定義の基準と標準に基づいて、書かれたテキストの品質を比較・評価します。
- [文法の魔法使い](https://docs.claude.com/ja/resources/prompt-library/grammar-genie.md): 文法的に間違った文章を正しい英語に変換します。
- [ユーモラスなヘルパーHal](https://docs.claude.com/ja/resources/prompt-library/hal-the-humorous-helper.md): 皮肉な一面を持つ知識豊富なAIとチャットしましょう。
- [慣用句解説者](https://docs.claude.com/ja/resources/prompt-library/idiom-illuminator.md): 一般的な慣用句やことわざの意味と由来を説明します。
- [面接質問作成者](https://docs.claude.com/ja/resources/prompt-library/interview-question-crafter.md): 面接用の質問を生成します。
- [LaTeX の達人](https://docs.claude.com/ja/resources/prompt-library/latex-legend.md): LaTeX文書を作成し、数式、表などのコードを生成します。
- [授業計画立案者](https://docs.claude.com/ja/resources/prompt-library/lesson-planner.md): あらゆる科目について詳細な授業計画を作成します。
- [プロンプトライブラリ](https://docs.claude.com/ja/resources/prompt-library/library.md)
- [マスターモデレーター](https://docs.claude.com/ja/resources/prompt-library/master-moderator.md): ユーザーの入力を潜在的に有害または違法なコンテンツについて評価します。
- [会議書記](https://docs.claude.com/ja/resources/prompt-library/meeting-scribe.md): 会議を議論トピック、重要なポイント、アクションアイテムを含む簡潔な要約に整理します。
- [メモマエストロ](https://docs.claude.com/ja/resources/prompt-library/memo-maestro.md): 重要なポイントに基づいて包括的な会社メモを作成します。
- [マインドフルネス・メンター](https://docs.claude.com/ja/resources/prompt-library/mindfulness-mentor.md): ユーザーをマインドフルネス・エクササイズとストレス軽減のためのテクニックを通してガイドします。
- [ムードカラライザー](https://docs.claude.com/ja/resources/prompt-library/mood-colorizer.md): ムードのテキスト記述を対応するHEXコードに変換します。
- [モチベーションの女神](https://docs.claude.com/ja/resources/prompt-library/motivational-muse.md): ユーザーの入力に基づいて、パーソナライズされたモチベーションメッセージと肯定的な言葉を提供します。
- [新語創造者](https://docs.claude.com/ja/resources/prompt-library/neologism-creator.md): ユーザーが提供する概念やアイデアに基づいて新しい単語を発明し、その定義を提供します。
- [視点の検討者](https://docs.claude.com/ja/resources/prompt-library/perspectives-ponderer.md): ユーザーが提供したトピックの長所と短所を検討します。
- [哲学的思索](https://docs.claude.com/ja/resources/prompt-library/philosophical-musings.md): 深い哲学的議論と思考実験に参加する。
- [PII除去ツール](https://docs.claude.com/ja/resources/prompt-library/pii-purifier.md): テキスト文書から個人識別情報（PII）を自動的に検出し、削除します。
- [多言語の超能力](https://docs.claude.com/ja/resources/prompt-library/polyglot-superpowers.md): 任意の言語から任意の言語にテキストを翻訳します。
- [ポートマンテュー詩人](https://docs.claude.com/ja/resources/prompt-library/portmanteau-poet.md): 2つの単語を組み合わせて、新しい意味のあるポートマンテューを作成します。
- [プロダクトネーミングプロ](https://docs.claude.com/ja/resources/prompt-library/product-naming-pro.md): 説明とキーワードから魅力的な製品名を作成します。
- [文章校正ツール](https://docs.claude.com/ja/resources/prompt-library/prose-polisher.md): 高度な校正技術と提案により、書かれたコンテンツを洗練し改善します。
- [駄洒落の達人](https://docs.claude.com/ja/resources/prompt-library/pun-dit.md): 与えられたトピックに基づいて巧妙な駄洒落と言葉遊びを生成します。
- [Pythonバグバスター](https://docs.claude.com/ja/resources/prompt-library/python-bug-buster.md): Pythonコードのバグを検出し修正します。
- [レビュー分類器](https://docs.claude.com/ja/resources/prompt-library/review-classifier.md): フィードバックを事前に指定されたタグと分類に分類します。
- [なぞなぞを出してください](https://docs.claude.com/ja/resources/prompt-library/riddle-me-this.md): なぞなぞを生成し、ユーザーを解答に導きます。
- [SF シナリオシミュレーター](https://docs.claude.com/ja/resources/prompt-library/sci-fi-scenario-simulator.md): ユーザーと様々なサイエンスフィクションシナリオ、および関連する課題と考慮事項について議論する。
- [小学2年生向け簡略化ツール](https://docs.claude.com/ja/resources/prompt-library/second-grade-simplifier.md): 複雑なテキストを若い学習者にとって理解しやすくします。
- [直喩の達人](https://docs.claude.com/ja/resources/prompt-library/simile-savant.md): 基本的な説明から直喩を生成します。
- [ソクラテス式賢者](https://docs.claude.com/ja/resources/prompt-library/socratic-sage.md): ユーザーが提供したトピックについてソクラテス式の対話を行います。
- [スプレッドシート魔術師](https://docs.claude.com/ja/resources/prompt-library/spreadsheet-sorcerer.md): 様々な種類のデータを含むCSVスプレッドシートを生成します。
- [SQLの魔術師](https://docs.claude.com/ja/resources/prompt-library/sql-sorcerer.md): 日常言語をSQLクエリに変換します。
- [ストーリーテリングの相棒](https://docs.claude.com/ja/resources/prompt-library/storytelling-sidekick.md): ユーザーと協力して魅力的なストーリーを作成し、プロットのひねりとキャラクター開発を提供します。
- [タイムトラベルコンサルタント](https://docs.claude.com/ja/resources/prompt-library/time-travel-consultant.md): ユーザーが仮想的なタイムトラベルシナリオとその影響をナビゲートするのを支援します。
- [早口言葉](https://docs.claude.com/ja/resources/prompt-library/tongue-twister.md): 挑戦的な早口言葉を作成します。
- [トリビアジェネレーター](https://docs.claude.com/ja/resources/prompt-library/trivia-generator.md): 幅広いトピックでトリビア問題を生成し、必要に応じてヒントを提供します。
- [ツイートのトーン検出器](https://docs.claude.com/ja/resources/prompt-library/tweet-tone-detector.md): ツイートの背後にあるトーンと感情を検出します。
- [VRフィットネス革新者](https://docs.claude.com/ja/resources/prompt-library/vr-fitness-innovator.md): バーチャルリアリティフィットネスゲームの創造的なアイデアをブレインストーミングします。
- [ウェブサイトウィザード](https://docs.claude.com/ja/resources/prompt-library/website-wizard.md): ユーザーの仕様に基づいてワンページウェブサイトを作成します。
- [クイックスタート](https://docs.claude.com/ja/resources/quickstarts.md): Claudeと Claude API を素早く始める
- [システムステータス](https://docs.claude.com/ja/resources/status.md)
- [API 키 가져오기](https://docs.claude.com/ko/api/admin-api/apikeys/get-api-key.md)
- [API 키 목록](https://docs.claude.com/ko/api/admin-api/apikeys/list-api-keys.md)
- [API 키 업데이트](https://docs.claude.com/ko/api/admin-api/apikeys/update-api-key.md)
- [Claude Code 사용 보고서 가져오기](https://docs.claude.com/ko/api/admin-api/claude-code/get-claude-code-usage-report.md): Retrieve daily aggregated usage metrics for Claude Code users.
Enables organizations to analyze developer productivity and build custom dashboards.
- [초대 생성](https://docs.claude.com/ko/api/admin-api/invites/create-invite.md)
- [초대 삭제](https://docs.claude.com/ko/api/admin-api/invites/delete-invite.md)
- [초대 가져오기](https://docs.claude.com/ko/api/admin-api/invites/get-invite.md)
- [초대 목록](https://docs.claude.com/ko/api/admin-api/invites/list-invites.md)
- [조직 정보 가져오기](https://docs.claude.com/ko/api/admin-api/organization/get-me.md)
- [비용 보고서 가져오기](https://docs.claude.com/ko/api/admin-api/usage-cost/get-cost-report.md)
- [Messages API에 대한 사용량 보고서 가져오기](https://docs.claude.com/ko/api/admin-api/usage-cost/get-messages-usage-report.md)
- [사용자 가져오기](https://docs.claude.com/ko/api/admin-api/users/get-user.md)
- [사용자 목록](https://docs.claude.com/ko/api/admin-api/users/list-users.md)
- [사용자 제거](https://docs.claude.com/ko/api/admin-api/users/remove-user.md)
- [사용자 업데이트](https://docs.claude.com/ko/api/admin-api/users/update-user.md)
- [워크스페이스 멤버 추가](https://docs.claude.com/ko/api/admin-api/workspace_members/create-workspace-member.md)
- [워크스페이스 멤버 삭제](https://docs.claude.com/ko/api/admin-api/workspace_members/delete-workspace-member.md)
- [워크스페이스 멤버 가져오기](https://docs.claude.com/ko/api/admin-api/workspace_members/get-workspace-member.md)
- [워크스페이스 멤버 목록 조회](https://docs.claude.com/ko/api/admin-api/workspace_members/list-workspace-members.md)
- [워크스페이스 멤버 업데이트](https://docs.claude.com/ko/api/admin-api/workspace_members/update-workspace-member.md)
- [워크스페이스 아카이브](https://docs.claude.com/ko/api/admin-api/workspaces/archive-workspace.md)
- [워크스페이스 생성](https://docs.claude.com/ko/api/admin-api/workspaces/create-workspace.md)
- [워크스페이스 가져오기](https://docs.claude.com/ko/api/admin-api/workspaces/get-workspace.md)
- [워크스페이스 목록](https://docs.claude.com/ko/api/admin-api/workspaces/list-workspaces.md)
- [워크스페이스 업데이트](https://docs.claude.com/ko/api/admin-api/workspaces/update-workspace.md)
- [Admin API 개요](https://docs.claude.com/ko/api/administration-api.md)
- [베타 헤더](https://docs.claude.com/ko/api/beta-headers.md): Claude API에서 베타 헤더 사용에 대한 문서
- [메시지 배치 취소](https://docs.claude.com/ko/api/canceling-message-batches.md): Batches may be canceled any time before processing ends. Once cancellation is initiated, the batch enters a `canceling` state, at which time the system may complete any in-progress, non-interruptible requests before finalizing cancellation.
The number of canceled requests is specified in `request_counts`. To determine which requests were canceled, check the individual results within the batch. Note that cancellation may not result in any canceled requests if they were non-interruptible.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Claude Code Analytics API](https://docs.claude.com/ko/api/claude-code-analytics-api.md): Claude Code Analytics Admin API를 사용하여 조직의 Claude Code 사용 분석 및 생산성 지표에 프로그래밍 방식으로 액세스하세요.
- [Amazon Bedrock API](https://docs.claude.com/ko/api/claude-on-amazon-bedrock.md): Anthropic의 Claude 모델이 이제 Amazon Bedrock을 통해 일반적으로 사용 가능합니다.
- [Vertex AI API](https://docs.claude.com/ko/api/claude-on-vertex-ai.md): Anthropic의 Claude 모델이 이제 [Vertex AI](https://cloud.google.com/vertex-ai)를 통해 일반적으로 사용 가능합니다.
- [클라이언트 SDK](https://docs.claude.com/ko/api/client-sdks.md): Claude API를 더 쉽게 사용할 수 있도록 여러 인기 언어로 클라이언트 라이브러리를 제공합니다.
- [메시지 배치 생성](https://docs.claude.com/ko/api/creating-message-batches.md): Send a batch of Message creation requests.
The Message Batches API can be used to process multiple Messages API requests at once. Once a Message Batch is created, it begins processing immediately. Batches can take up to 24 hours to complete.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [메시지 배치 삭제](https://docs.claude.com/ko/api/deleting-message-batches.md): Delete a Message Batch.
Message Batches can only be deleted once they've finished processing. If you'd like to delete an in-progress batch, you must first cancel it.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [오류](https://docs.claude.com/ko/api/errors.md): API 오류 코드, 요청 크기 제한, 오류 형식 및 긴 요청 처리에 대한 가이드
- [파일 다운로드](https://docs.claude.com/ko/api/files-content.md): Claude가 생성한 파일의 내용을 다운로드합니다
- [파일 생성](https://docs.claude.com/ko/api/files-create.md): 파일 업로드
- [파일 삭제](https://docs.claude.com/ko/api/files-delete.md): API를 통해 파일에 액세스할 수 없도록 만들기
- [파일 목록](https://docs.claude.com/ko/api/files-list.md): 워크스페이스 내의 파일 목록을 조회합니다
- [파일 메타데이터 가져오기](https://docs.claude.com/ko/api/files-metadata.md): Claude API와 함께 사용할 파일을 업로드하고 관리할 수 있는 Files API에 대한 정보입니다.
- [도움 받기](https://docs.claude.com/ko/api/getting-help.md): 이 문서에서 가장 일반적인 질문에 대한 답변을 제공하려고 노력했습니다. 그러나 Claude, Claude API 또는 당사의 제품 사용에 대한 추가 기술 지원이 필요한 경우, [support.claude.com](https://support.claude.com)에서 지원팀에 문의할 수 있습니다.
- [중단 이유 처리](https://docs.claude.com/ko/api/handling-stop-reasons.md): Claude의 응답에서 stop_reason 필드를 이해하고 처리하는 방법을 알아보세요.
- [IP 주소](https://docs.claude.com/ko/api/ip-addresses.md): Anthropic 서비스는 인바운드 및 아웃바운드 연결 모두에 고정 IP 주소를 사용합니다. 이러한 주소를 사용하여 Claude API 및 Console에 대한 보안 액세스를 위한 방화벽 규칙을 구성할 수 있습니다. 이러한 주소는 사전 통지 없이 변경되지 않습니다.
- [메시지 배치 목록](https://docs.claude.com/ko/api/listing-message-batches.md): List all Message Batches within a Workspace. Most recently created batches are returned first.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [메시지](https://docs.claude.com/ko/api/messages.md): Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless multi-turn conversations.
Learn more about the Messages API in our [user guide](/en/docs/initial-setup)
- [Message Batches 예제](https://docs.claude.com/ko/api/messages-batch-examples.md): Message Batches API 사용 예제
- [메시지 토큰 개수 세기](https://docs.claude.com/ko/api/messages-count-tokens.md): Count the number of tokens in a Message.
The Token Count API can be used to count the number of tokens in a Message, including tools, images, and documents, without creating it.
Learn more about token counting in our [user guide](/en/docs/build-with-claude/token-counting)
- [Messages 예제](https://docs.claude.com/ko/api/messages-examples.md): Messages API에 대한 요청 및 응답 예제
- [Text Completions에서 마이그레이션](https://docs.claude.com/ko/api/migrating-from-text-completions-to-messages.md): Text Completions에서 Messages로 마이그레이션
- [모델 가져오기](https://docs.claude.com/ko/api/models.md): Get a specific model.
The Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.
- [모델 목록](https://docs.claude.com/ko/api/models-list.md): List available models.
The Models API response can be used to determine which models are available for use in the API. More recently released models are listed first.
- [OpenAI SDK 호환성](https://docs.claude.com/ko/api/openai-sdk.md): Anthropic은 OpenAI SDK를 사용하여 Claude API를 테스트할 수 있는 호환성 레이어를 제공합니다. 몇 가지 코드 변경으로 Anthropic 모델 기능을 빠르게 평가할 수 있습니다.
- [개요](https://docs.claude.com/ko/api/overview.md): Claude API에 대한 개요
- [프롬프트 생성](https://docs.claude.com/ko/api/prompt-tools-generate.md): 잘 작성된 프롬프트 생성
- [프롬프트 개선하기](https://docs.claude.com/ko/api/prompt-tools-improve.md): 피드백을 바탕으로 새롭고 개선된 프롬프트 생성하기
- [프롬프트 템플릿화](https://docs.claude.com/ko/api/prompt-tools-templatize.md): 변수를 식별하고 추출하여 프롬프트를 템플릿화합니다
- [속도 제한](https://docs.claude.com/ko/api/rate-limits.md): API의 오남용을 방지하고 용량을 관리하기 위해, 조직이 Claude API를 사용할 수 있는 양에 대한 제한을 구현했습니다.
- [메시지 배치 결과 검색](https://docs.claude.com/ko/api/retrieving-message-batch-results.md): Streams the results of a Message Batch as a `.jsonl` file.
Each line in the file is a JSON object containing the result of a single request in the Message Batch. Results are not guaranteed to be in the same order as requests. Use the `custom_id` field to match results to requests.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [메시지 배치 검색](https://docs.claude.com/ko/api/retrieving-message-batches.md): This endpoint is idempotent and can be used to poll for Message Batch completion. To access the results of a Message Batch, make a request to the `results_url` field in the response.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [서비스 티어](https://docs.claude.com/ko/api/service-tiers.md): 다양한 서비스 티어를 통해 애플리케이션의 요구사항에 따라 가용성, 성능, 예측 가능한 비용의 균형을 맞출 수 있습니다.
- [지원되는 지역](https://docs.claude.com/ko/api/supported-regions.md): 현재 액세스를 지원할 수 있는 국가, 지역 및 영토는 다음과 같습니다:
- [사용량 및 비용 API](https://docs.claude.com/ko/api/usage-cost-api.md): 사용량 및 비용 관리자 API를 통해 조직의 API 사용량 및 비용 데이터에 프로그래밍 방식으로 액세스하세요.
- [버전](https://docs.claude.com/ko/api/versioning.md): API 요청을 할 때 `anthropic-version` 요청 헤더를 보내야 합니다. 예를 들어, `anthropic-version: 2023-06-01`입니다. [클라이언트 SDK](/ko/api/client-sdks)를 사용하는 경우 이는 자동으로 처리됩니다.
- [용어집](https://docs.claude.com/ko/docs/about-claude/glossary.md): 이러한 개념들은 Anthropic의 언어 모델에만 국한되지 않지만, 아래에 주요 용어들에 대한 간략한 요약을 제시합니다.
- [모델 지원 중단](https://docs.claude.com/ko/docs/about-claude/model-deprecations.md)
- [적합한 모델 선택하기](https://docs.claude.com/ko/docs/about-claude/models/choosing-a-model.md): 애플리케이션에 최적의 Claude 모델을 선택하는 것은 기능, 속도, 비용이라는 세 가지 주요 고려사항의 균형을 맞추는 것입니다. 이 가이드는 특정 요구사항에 따라 정보에 입각한 결정을 내리는 데 도움을 줍니다.
- [Claude 4로 마이그레이션](https://docs.claude.com/ko/docs/about-claude/models/migrating-to-claude-4.md): Claude 3.7 모델에서 Claude 4 모델로 마이그레이션하는 방법에 대한 가이드입니다.
- [모델 개요](https://docs.claude.com/ko/docs/about-claude/models/overview.md): Claude는 Anthropic에서 개발한 최첨단 대형 언어 모델 제품군입니다. 이 가이드는 저희 모델을 소개하고 레거시 모델과의 성능을 비교합니다.
- [가격](https://docs.claude.com/ko/docs/about-claude/pricing.md): 모델과 기능에 대한 Anthropic의 가격 구조에 대해 알아보세요
- [콘텐츠 조정](https://docs.claude.com/ko/docs/about-claude/use-case-guides/content-moderation.md): 콘텐츠 조정은 디지털 애플리케이션에서 안전하고 존중받으며 생산적인 환경을 유지하는 데 중요한 측면입니다. 이 가이드에서는 디지털 애플리케이션 내에서 콘텐츠를 조정하는 데 Claude를 사용하는 방법에 대해 논의하겠습니다.
- [고객 지원 에이전트](https://docs.claude.com/ko/docs/about-claude/use-case-guides/customer-support-chat.md): 이 가이드는 Claude의 고급 대화 기능을 활용하여 실시간으로 고객 문의를 처리하고, 24/7 지원을 제공하며, 대기 시간을 줄이고, 정확한 응답과 긍정적인 상호작용으로 높은 지원 볼륨을 관리하는 방법을 안내합니다.
- [법률 문서 요약](https://docs.claude.com/ko/docs/about-claude/use-case-guides/legal-summarization.md): 이 가이드는 Claude의 고급 자연어 처리 기능을 활용하여 법률 문서를 효율적으로 요약하고, 핵심 정보를 추출하며, 법률 연구를 가속화하는 방법을 안내합니다. Claude를 통해 계약서 검토, 소송 준비, 규제 업무를 간소화하여 시간을 절약하고 법률 프로세스의 정확성을 보장할 수 있습니다.
- [일반적인 사용 사례 가이드](https://docs.claude.com/ko/docs/about-claude/use-case-guides/overview.md)
- [티켓 라우팅](https://docs.claude.com/ko/docs/about-claude/use-case-guides/ticket-routing.md): 이 가이드는 Claude의 고급 자연어 이해 능력을 활용하여 고객 의도, 긴급성, 우선순위, 고객 프로필 등을 기반으로 고객 지원 티켓을 대규모로 분류하는 방법을 안내합니다.
- [Google Sheets 애드온](https://docs.claude.com/ko/docs/agents-and-tools/claude-for-sheets.md): [Claude for Sheets 확장 프로그램](https://workspace.google.com/marketplace/app/claude%5Ffor%5Fsheets/909417792257)은 Claude를 Google Sheets에 통합하여 셀에서 직접 Claude와 상호작용할 수 있게 해줍니다.
- [MCP 커넥터](https://docs.claude.com/ko/docs/agents-and-tools/mcp-connector.md): Claude의 Model Context Protocol (MCP) 커넥터 기능을 사용하여 별도의 MCP 클라이언트 없이 Messages API에서 직접 원격 MCP 서버에 연결하는 방법을 알아보세요.
- [원격 MCP 서버](https://docs.claude.com/ko/docs/agents-and-tools/remote-mcp-servers.md): 원격 MCP 서버에 대한 정보
- [Bash 도구](https://docs.claude.com/ko/docs/agents-and-tools/tool-use/bash-tool.md): Claude가 지속적인 bash 세션에서 셸 명령을 실행할 수 있게 하는 bash 도구에 대해 알아보세요.
- [코드 실행 도구](https://docs.claude.com/ko/docs/agents-and-tools/tool-use/code-execution-tool.md)
- [컴퓨터 사용 도구](https://docs.claude.com/ko/docs/agents-and-tools/tool-use/computer-use-tool.md): Claude가 컴퓨터 환경과 상호작용할 수 있게 해주는 컴퓨터 사용 도구에 대해 알아보세요. 스크린샷 캡처와 마우스/키보드 제어를 통한 자율적인 데스크톱 상호작용을 제공합니다.
- [세밀한 도구 스트리밍](https://docs.claude.com/ko/docs/agents-and-tools/tool-use/fine-grained-tool-streaming.md): 도구 사용에서 매개변수 값에 대한 세밀한 스트리밍을 지원하여 대용량 매개변수 수신 시작 지연 시간을 줄입니다.
- [도구 사용 구현 방법](https://docs.claude.com/ko/docs/agents-and-tools/tool-use/implement-tool-use.md): Claude에서 도구 사용을 구현하는 방법에 대한 포괄적인 가이드입니다.
- [Claude와 함께하는 도구 사용](https://docs.claude.com/ko/docs/agents-and-tools/tool-use/overview.md)
- [텍스트 에디터 도구](https://docs.claude.com/ko/docs/agents-and-tools/tool-use/text-editor-tool.md): Claude가 텍스트 파일을 보고 수정할 수 있는 Anthropic 정의 텍스트 에디터 도구를 사용하여 코드나 기타 텍스트 문서를 디버그, 수정 및 개선하는 방법을 알아보세요.
- [토큰 효율적인 도구 사용](https://docs.claude.com/ko/docs/agents-and-tools/tool-use/token-efficient-tool-use.md)
- [웹 페치 도구](https://docs.claude.com/ko/docs/agents-and-tools/tool-use/web-fetch-tool.md): Claude가 지정된 웹 페이지와 PDF 문서에서 전체 콘텐츠를 검색할 수 있게 해주는 웹 페치 도구입니다.
- [웹 검색 도구](https://docs.claude.com/ko/docs/agents-and-tools/tool-use/web-search-tool.md)
- [배치 처리](https://docs.claude.com/ko/docs/build-with-claude/batch-processing.md)
- [인용](https://docs.claude.com/ko/docs/build-with-claude/citations.md): Claude가 문서에 대한 질문에 답할 때 상세한 인용을 제공하여 응답에서 정보 출처를 추적하고 검증할 수 있도록 도와줍니다.
- [컨텍스트 윈도우](https://docs.claude.com/ko/docs/build-with-claude/context-windows.md)
- [임베딩](https://docs.claude.com/ko/docs/build-with-claude/embeddings.md): 텍스트 임베딩은 의미적 유사성을 측정할 수 있게 해주는 텍스트의 수치적 표현입니다. 이 가이드는 임베딩, 그 응용 분야, 그리고 검색, 추천, 이상 탐지와 같은 작업에 임베딩 모델을 사용하는 방법을 소개합니다.
- [확장된 사고로 구축하기](https://docs.claude.com/ko/docs/build-with-claude/extended-thinking.md)
- [Files API](https://docs.claude.com/ko/docs/build-with-claude/files.md)
- [다국어 지원](https://docs.claude.com/ko/docs/build-with-claude/multilingual-support.md): Claude는 여러 언어에 걸친 작업에 탁월하며, 영어에 비해 강력한 교차 언어 성능을 유지합니다.
- [기능 개요](https://docs.claude.com/ko/docs/build-with-claude/overview.md): Claude의 고급 기능과 역량을 살펴보세요.
- [PDF 지원](https://docs.claude.com/ko/docs/build-with-claude/pdf-support.md): Claude로 PDF를 처리하세요. 문서에서 텍스트를 추출하고, 차트를 분석하며, 시각적 콘텐츠를 이해할 수 있습니다.
- [프롬프트 캐싱](https://docs.claude.com/ko/docs/build-with-claude/prompt-caching.md)
- [명확하고 직접적이며 상세하게](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/be-clear-and-direct.md)
- [Claude가 생각하도록 하여(사고 연쇄 프롬프팅) 성능을 향상시키기](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/chain-of-thought.md)
- [더 강력한 성능을 위한 복잡한 프롬프트 체이닝](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/chain-prompts.md)
- [Claude 4 프롬프트 엔지니어링 모범 사례](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/claude-4-best-practices.md): Claude 4 모델을 위한 특정 프롬프트 엔지니어링 기법을 제공하여 애플리케이션에서 최적의 결과를 달성할 수 있도록 도와줍니다.
- [확장 사고 팁](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/extended-thinking-tips.md): Claude의 확장 사고 기능을 최대한 활용하기 위한 고급 전략과 기법을 제공합니다.
- [긴 컨텍스트 프롬프팅 팁](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/long-context-tips.md)
- [예시(멀티샷 프롬프팅)를 사용하여 Claude의 행동 유도하기](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/multishot-prompting.md)
- [프롬프트 엔지니어링 개요](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/overview.md)
- [더 나은 출력 제어를 위한 Claude의 응답 미리 채우기](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/prefill-claudes-response.md)
- [첫 초안 프롬프트 템플릿 자동 생성](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/prompt-generator.md)
- [프롬프트 개선기를 사용하여 프롬프트 최적화하기](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/prompt-improver.md): 프롬프트 개선기를 사용하여 프롬프트를 최적화하는 방법을 알아보세요
- [프롬프트 템플릿과 변수 사용하기](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/prompt-templates-and-variables.md)
- [시스템 프롬프트로 Claude에게 역할 부여하기](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/system-prompts.md)
- [XML 태그를 사용하여 프롬프트 구조화하기](https://docs.claude.com/ko/docs/build-with-claude/prompt-engineering/use-xml-tags.md)
- [검색 결과](https://docs.claude.com/ko/docs/build-with-claude/search-results.md): 소스 출처가 포함된 검색 결과를 제공하여 RAG 애플리케이션에서 자연스러운 인용을 가능하게 합니다
- [스트리밍 메시지](https://docs.claude.com/ko/docs/build-with-claude/streaming.md): 서버 전송 이벤트를 사용하여 메시지 응답을 점진적으로 스트리밍하는 방법을 알아보세요.
- [토큰 카운팅](https://docs.claude.com/ko/docs/build-with-claude/token-counting.md): Claude에게 메시지를 보내기 전에 메시지의 토큰 수를 결정하여 프롬프트와 사용량에 대한 정보에 입각한 결정을 내릴 수 있도록 도와줍니다.
- [비전](https://docs.claude.com/ko/docs/build-with-claude/vision.md): Claude 3 및 4 모델 패밀리는 Claude가 이미지를 이해하고 분석할 수 있게 해주는 새로운 비전 기능을 제공하여 멀티모달 상호작용의 흥미로운 가능성을 열어줍니다.
- [Amazon Bedrock에서 Claude Code](https://docs.claude.com/ko/docs/claude-code/amazon-bedrock.md): 설정, IAM 구성 및 문제 해결을 포함하여 Amazon Bedrock을 통해 Claude Code를 구성하는 방법을 알아보세요.
- [분석](https://docs.claude.com/ko/docs/claude-code/analytics.md): 조직의 Claude Code 배포에 대한 자세한 사용량 인사이트와 생산성 메트릭을 확인하세요.
- [CLI 참조](https://docs.claude.com/ko/docs/claude-code/cli-reference.md): Claude Code 명령줄 인터페이스의 완전한 참조, 명령어와 플래그 포함.
- [일반적인 워크플로우](https://docs.claude.com/ko/docs/claude-code/common-workflows.md): Claude Code를 사용한 일반적인 워크플로우에 대해 알아보세요.
- [비용을 효과적으로 관리하기](https://docs.claude.com/ko/docs/claude-code/costs.md): Claude Code를 사용할 때 토큰 사용량과 비용을 추적하고 최적화하는 방법을 알아보세요.
- [데이터 사용](https://docs.claude.com/ko/docs/claude-code/data-usage.md): Claude에 대한 Anthropic의 데이터 사용 정책에 대해 알아보세요
- [개발 컨테이너](https://docs.claude.com/ko/docs/claude-code/devcontainer.md): 일관되고 안전한 환경이 필요한 팀을 위한 Claude Code 개발 컨테이너에 대해 알아보세요.
- [Claude Code GitHub Actions](https://docs.claude.com/ko/docs/claude-code/github-actions.md): Claude Code GitHub Actions를 사용하여 개발 워크플로우에 Claude Code를 통합하는 방법을 알아보세요
- [Claude Code GitLab CI/CD](https://docs.claude.com/ko/docs/claude-code/gitlab-ci-cd.md): GitLab CI/CD를 사용하여 Claude Code를 개발 워크플로우에 통합하는 방법을 알아보세요
- [Google Vertex AI의 Claude Code](https://docs.claude.com/ko/docs/claude-code/google-vertex-ai.md): 설정, IAM 구성, 문제 해결을 포함하여 Google Vertex AI를 통한 Claude Code 구성에 대해 알아보세요.
- [훅 참조](https://docs.claude.com/ko/docs/claude-code/hooks.md): 이 페이지는 Claude Code에서 훅을 구현하기 위한 참조 문서를 제공합니다.
- [Claude Code 훅 시작하기](https://docs.claude.com/ko/docs/claude-code/hooks-guide.md): 셸 명령어를 등록하여 Claude Code의 동작을 사용자 정의하고 확장하는 방법을 알아보세요
- [신원 및 접근 관리](https://docs.claude.com/ko/docs/claude-code/iam.md): 조직에서 Claude Code의 사용자 인증, 권한 부여 및 접근 제어를 구성하는 방법을 알아보세요.
- [IDE에 Claude Code 추가하기](https://docs.claude.com/ko/docs/claude-code/ide-integrations.md): 좋아하는 IDE에 Claude Code를 추가하는 방법을 알아보세요
- [대화형 모드](https://docs.claude.com/ko/docs/claude-code/interactive-mode.md): Claude Code 세션에서 키보드 단축키, 입력 모드 및 대화형 기능에 대한 완전한 참조.
- [법적 및 규정 준수](https://docs.claude.com/ko/docs/claude-code/legal-and-compliance.md): Claude Code에 대한 법적 계약, 규정 준수 인증 및 보안 정보입니다.
- [LLM 게이트웨이 구성](https://docs.claude.com/ko/docs/claude-code/llm-gateway.md): LiteLLM 설정, 인증 방법, 사용량 추적 및 예산 관리와 같은 엔터프라이즈 기능을 포함하여 LLM 게이트웨이 솔루션으로 Claude Code를 구성하는 방법을 알아보세요.
- [MCP를 통해 Claude Code를 도구에 연결하기](https://docs.claude.com/ko/docs/claude-code/mcp.md): Model Context Protocol을 사용하여 Claude Code를 도구에 연결하는 방법을 알아보세요.
- [Claude의 메모리 관리](https://docs.claude.com/ko/docs/claude-code/memory.md): 다양한 메모리 위치와 모범 사례를 통해 세션 간 Claude Code의 메모리를 관리하는 방법을 알아보세요.
- [모델 구성](https://docs.claude.com/ko/docs/claude-code/model-config.md): Claude Code 모델 구성에 대해 알아보세요. `opusplan`과 같은 모델 별칭을 포함합니다.
- [모니터링](https://docs.claude.com/ko/docs/claude-code/monitoring-usage.md): Claude Code에서 OpenTelemetry를 활성화하고 구성하는 방법을 알아보세요.
- [엔터프라이즈 네트워크 구성](https://docs.claude.com/ko/docs/claude-code/network-config.md): 프록시 서버, 사용자 정의 인증 기관(CA), 상호 전송 계층 보안(mTLS) 인증을 사용하는 엔터프라이즈 환경에서 Claude Code를 구성합니다.
- [출력 스타일](https://docs.claude.com/ko/docs/claude-code/output-styles.md): 소프트웨어 엔지니어링을 넘어선 용도로 Claude Code를 적응시키기
- [Claude Code 개요](https://docs.claude.com/ko/docs/claude-code/overview.md): 터미널에서 작동하며 아이디어를 그 어느 때보다 빠르게 코드로 변환하는 데 도움을 주는 Anthropic의 에이전틱 코딩 도구인 Claude Code에 대해 알아보세요.
- [빠른 시작](https://docs.claude.com/ko/docs/claude-code/quickstart.md): Claude Code에 오신 것을 환영합니다!
- [커스텀 도구](https://docs.claude.com/ko/docs/claude-code/sdk/custom-tools.md): Claude Code SDK 기능을 확장하기 위한 커스텀 도구를 구축하고 통합하세요
- [시스템 프롬프트 수정하기](https://docs.claude.com/ko/docs/claude-code/sdk/modifying-system-prompts.md): 출력 스타일, appendSystemPrompt, customSystemPrompt 세 가지 접근 방식을 사용하여 시스템 프롬프트를 수정해 Claude의 동작을 사용자 정의하는 방법을 알아보세요.
- [비용 및 사용량 추적](https://docs.claude.com/ko/docs/claude-code/sdk/sdk-cost-tracking.md): Claude Code SDK에서 청구를 위한 토큰 사용량을 이해하고 추적하기
- [헤드리스 모드](https://docs.claude.com/ko/docs/claude-code/sdk/sdk-headless.md): 대화형 UI 없이 프로그래밍 방식으로 Claude Code 실행
- [SDK에서의 MCP](https://docs.claude.com/ko/docs/claude-code/sdk/sdk-mcp.md): Model Context Protocol 서버를 사용하여 사용자 정의 도구로 Claude Code를 확장하세요
- [개요](https://docs.claude.com/ko/docs/claude-code/sdk/sdk-overview.md): Claude Code SDK로 맞춤형 AI 에이전트 구축하기
- [권한 처리](https://docs.claude.com/ko/docs/claude-code/sdk/sdk-permissions.md): Claude Code SDK에서 도구 사용 및 권한을 제어하세요
- [Python SDK 참조](https://docs.claude.com/ko/docs/claude-code/sdk/sdk-python.md): 모든 함수, 타입 및 클래스를 포함한 Claude Code Python SDK의 완전한 API 참조입니다.
- [세션 관리](https://docs.claude.com/ko/docs/claude-code/sdk/sdk-sessions.md): Claude Code SDK가 세션과 세션 재개를 처리하는 방법 이해하기
- [SDK의 슬래시 명령어](https://docs.claude.com/ko/docs/claude-code/sdk/sdk-slash-commands.md): SDK를 통해 Claude Code 세션을 제어하기 위한 슬래시 명령어 사용법을 알아보세요
- [TypeScript SDK 참조](https://docs.claude.com/ko/docs/claude-code/sdk/sdk-typescript.md): 모든 함수, 타입, 인터페이스를 포함한 Claude Code TypeScript SDK의 완전한 API 참조입니다.
- [스트리밍 입력](https://docs.claude.com/ko/docs/claude-code/sdk/streaming-vs-single-mode.md): Claude Code SDK의 두 가지 입력 모드와 각각을 언제 사용해야 하는지 이해하기
- [SDK의 서브에이전트](https://docs.claude.com/ko/docs/claude-code/sdk/subagents.md): Claude Code SDK에서 서브에이전트 작업하기
- [할 일 목록](https://docs.claude.com/ko/docs/claude-code/sdk/todo-tracking.md): 체계적인 작업 관리를 위해 Claude Code SDK를 사용하여 할 일을 추적하고 표시합니다
- [보안](https://docs.claude.com/ko/docs/claude-code/security.md): Claude Code의 보안 보호 장치와 안전한 사용을 위한 모범 사례에 대해 알아보세요.
- [Claude Code 설정](https://docs.claude.com/ko/docs/claude-code/settings.md): 전역 및 프로젝트 수준 설정과 환경 변수로 Claude Code를 구성하세요.
- [Claude Code 설정](https://docs.claude.com/ko/docs/claude-code/setup.md): 개발 머신에서 Claude Code를 설치, 인증 및 사용을 시작하세요.
- [슬래시 명령어](https://docs.claude.com/ko/docs/claude-code/slash-commands.md): 대화형 세션에서 슬래시 명령어로 Claude의 동작을 제어하세요.
- [상태 표시줄 구성](https://docs.claude.com/ko/docs/claude-code/statusline.md): Claude Code에서 상황별 정보를 표시하는 사용자 정의 상태 표시줄을 만들어보세요
- [서브에이전트](https://docs.claude.com/ko/docs/claude-code/sub-agents.md): Claude Code에서 작업별 워크플로우와 향상된 컨텍스트 관리를 위한 전문 AI 서브에이전트를 생성하고 사용하세요.
- [터미널 설정 최적화](https://docs.claude.com/ko/docs/claude-code/terminal-config.md): Claude Code는 터미널이 올바르게 구성되었을 때 최상의 성능을 발휘합니다. 이 가이드라인을 따라 경험을 최적화하세요.
- [엔터프라이즈 배포 개요](https://docs.claude.com/ko/docs/claude-code/third-party-integrations.md): Claude Code가 다양한 타사 서비스 및 인프라와 통합하여 엔터프라이즈 배포 요구사항을 충족하는 방법을 알아보세요.
- [문제 해결](https://docs.claude.com/ko/docs/claude-code/troubleshooting.md): Claude Code 설치 및 사용 시 발생하는 일반적인 문제에 대한 해결책을 알아보세요.
- [Claude 시작하기](https://docs.claude.com/ko/docs/get-started.md): Claude에 첫 번째 API 호출을 하고 간단한 웹 검색 어시스턴트를 구축하세요
- [Claude 소개](https://docs.claude.com/ko/docs/intro.md): Claude는 Anthropic에서 구축한 고성능, 신뢰할 수 있고 지능적인 AI 플랫폼입니다. Claude는 언어, 추론, 분석, 코딩 등과 관련된 작업에서 뛰어난 성능을 발휘합니다.
- [Anthropic 개인정보 보호정책](https://docs.claude.com/ko/docs/legal-center/privacy.md)
- [보안 및 규정 준수](https://docs.claude.com/ko/docs/legal-center/security-compliance.md)
- [Claude로 구축하기](https://docs.claude.com/ko/docs/overview.md): Claude는 Anthropic에서 구축한 [고성능이고 지능적인 AI 모델](/ko/docs/about-claude/models) 제품군입니다. Claude는 강력하고 확장 가능하면서도 가장 신뢰할 수 있고 안정적인 AI입니다. 중요한 프로토콜을 따르고, 실수를 적게 하며, 탈옥에 저항력이 있어 [기업 고객](https://claude.com/customers)이 대규모로 가장 안전한 AI 기반 애플리케이션을 구축할 수 있게 합니다.
- [성공 기준 정의하기](https://docs.claude.com/ko/docs/test-and-evaluate/define-success.md): 성공적인 LLM 기반 애플리케이션을 구축하기 위한 명확한 성공 기준을 정의하는 방법
- [강력한 실증적 평가 만들기](https://docs.claude.com/ko/docs/test-and-evaluate/develop-tests.md): 성공 기준을 정의한 후, 다음 단계는 해당 기준에 대한 LLM 성능을 측정하는 평가를 설계하는 것입니다. 이는 프롬프트 엔지니어링 사이클의 중요한 부분입니다.
- [평가 도구 사용하기](https://docs.claude.com/ko/docs/test-and-evaluate/eval-tool.md): [Claude Console](https://console.anthropic.com/dashboard)은 다양한 시나리오에서 프롬프트를 테스트할 수 있는 **평가 도구**를 제공합니다.
- [null](https://docs.claude.com/ko/docs/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals.md)
- [출력 일관성 높이기 (JSON 모드)](https://docs.claude.com/ko/docs/test-and-evaluate/strengthen-guardrails/increase-consistency.md)
- [역할 프롬프팅과 사전 채우기로 Claude의 캐릭터 유지하기](https://docs.claude.com/ko/docs/test-and-evaluate/strengthen-guardrails/keep-claude-in-character.md)
- [탈옥 및 프롬프트 인젝션 완화](https://docs.claude.com/ko/docs/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks.md)
- [환각 현상 줄이기](https://docs.claude.com/ko/docs/test-and-evaluate/strengthen-guardrails/reduce-hallucinations.md)
- [지연 시간 감소](https://docs.claude.com/ko/docs/test-and-evaluate/strengthen-guardrails/reduce-latency.md)
- [프롬프트 유출 감소](https://docs.claude.com/ko/docs/test-and-evaluate/strengthen-guardrails/reduce-prompt-leak.md)
- [null](https://docs.claude.com/ko/home.md)
- [Claude 개발자 플랫폼](https://docs.claude.com/ko/release-notes/api.md): Claude API와 Claude Console의 업데이트를 따라가세요.
- [Claude Apps](https://docs.claude.com/ko/release-notes/claude-apps.md): Anthropic의 Claude 애플리케이션 전반의 업데이트를 따라가 보세요.
- [Claude Code](https://docs.claude.com/ko/release-notes/claude-code.md): Claude Code의 최신 기능과 개선 사항을 최신 상태로 유지하세요
- [개요](https://docs.claude.com/ko/release-notes/overview.md): Anthropic의 제품과 서비스 전반의 업데이트를 따라가 보세요.
- [시스템 프롬프트](https://docs.claude.com/ko/release-notes/system-prompts.md): [Claude.ai](https://www.claude.ai)와 Claude [iOS](http://anthropic.com/ios) 및 [Android](http://anthropic.com/android) 앱의 핵심 시스템 프롬프트 업데이트를 확인하세요.
- [Claude 3.7 시스템 카드](https://docs.claude.com/ko/resources/claude-3-7-system-card.md)
- [Claude 3 모델 카드](https://docs.claude.com/ko/resources/claude-3-model-card.md)
- [Claude 4 시스템 카드](https://docs.claude.com/ko/resources/claude-4-system-card.md)
- [Anthropic 코스](https://docs.claude.com/ko/resources/courses.md)
- [null](https://docs.claude.com/ko/resources/overview.md)
- [적응형 편집기](https://docs.claude.com/ko/resources/prompt-library/adaptive-editor.md): 다른 톤, 대상, 또는 스타일과 같이 사용자가 제공한 지침에 따라 텍스트를 다시 작성합니다.
- [공항 코드 분석기](https://docs.claude.com/ko/resources/prompt-library/airport-code-analyst.md): 텍스트에서 공항 코드를 찾고 추출합니다.
- [외계 인류학자](https://docs.claude.com/ko/resources/prompt-library/alien-anthropologist.md): 외계 인류학자의 관점에서 인간 문화와 관습을 분석합니다.
- [두운 연금술사](https://docs.claude.com/ko/resources/prompt-library/alliteration-alchemist.md): 주어진 주제에 대한 두운 구문과 문장을 생성합니다.
- [바벨의 방송](https://docs.claude.com/ko/resources/prompt-library/babels-broadcasts.md): 세계에서 가장 많이 사용되는 10개 언어로 매력적인 제품 발표 트윗을 작성하세요.
- [브랜드 빌더](https://docs.claude.com/ko/resources/prompt-library/brand-builder.md): 전체적인 브랜드 아이덴티티를 위한 디자인 브리프를 작성합니다.
- [커리어 코치](https://docs.claude.com/ko/resources/prompt-library/career-coach.md): AI 커리어 코치와 역할극 대화를 나누세요.
- [출처 인용하기](https://docs.claude.com/ko/resources/prompt-library/cite-your-sources.md): 응답을 뒷받침하는 관련 인용과 함께 문서 내용에 대한 질문에 대한 답변을 얻으세요.
- [코드 설명기](https://docs.claude.com/ko/resources/prompt-library/code-clarifier.md): 복잡한 코드를 평이한 언어로 단순화하고 설명합니다.
- [코드 컨설턴트](https://docs.claude.com/ko/resources/prompt-library/code-consultant.md): Python 코드 성능을 최적화하기 위한 개선 사항을 제안합니다.
- [기업 투시자](https://docs.claude.com/ko/resources/prompt-library/corporate-clairvoyant.md): 긴 기업 보고서에서 통찰력을 추출하고, 위험을 식별하며, 핵심 정보를 하나의 메모로 요약합니다.
- [우주 키스트로크](https://docs.claude.com/ko/resources/prompt-library/cosmic-keystrokes.md): 사이드 스크롤링 게임플레이와 Tailwind CSS 스타일링을 특징으로 하는 인터랙티브 속도 타이핑 게임을 단일 HTML 파일로 생성합니다.
- [CSV 변환기](https://docs.claude.com/ko/resources/prompt-library/csv-converter.md): 다양한 형식(JSON, XML 등)의 데이터를 적절히 형식화된 CSV 파일로 변환합니다.
- [요리 창작자](https://docs.claude.com/ko/resources/prompt-library/culinary-creator.md): 사용자의 사용 가능한 재료와 식단 선호도를 바탕으로 레시피 아이디어를 제안합니다.
- [데이터 정리기](https://docs.claude.com/ko/resources/prompt-library/data-organizer.md): 비구조화된 텍스트를 맞춤형 JSON 테이블로 변환합니다.
- [방향 디코더](https://docs.claude.com/ko/resources/prompt-library/direction-decoder.md): 자연어를 단계별 지침으로 변환합니다.
- [꿈 해석사](https://docs.claude.com/ko/resources/prompt-library/dream-interpreter.md): 사용자의 꿈에 나타나는 상징성에 대한 해석과 통찰을 제공합니다.
- [효율성 추정기](https://docs.claude.com/ko/resources/prompt-library/efficiency-estimator.md): 함수와 알고리즘의 시간 복잡도를 계산합니다.
- [이메일 추출기](https://docs.claude.com/ko/resources/prompt-library/email-extractor.md): 문서에서 이메일 주소를 추출하여 JSON 형식의 목록으로 만듭니다.
- [이모지 인코더](https://docs.claude.com/ko/resources/prompt-library/emoji-encoder.md): 평범한 텍스트를 재미있고 표현력 있는 이모지 메시지로 변환합니다.
- [윤리적 딜레마 내비게이터](https://docs.claude.com/ko/resources/prompt-library/ethical-dilemma-navigator.md): 사용자가 복잡한 윤리적 딜레마를 생각해보고 다양한 관점을 제공하도록 도움을 줍니다.
- [Excel 수식 전문가](https://docs.claude.com/ko/resources/prompt-library/excel-formula-expert.md): 사용자가 설명한 계산이나 데이터 조작을 기반으로 Excel 수식을 생성합니다.
- [함수 제작자](https://docs.claude.com/ko/resources/prompt-library/function-fabricator.md): 상세한 명세를 기반으로 Python 함수를 생성합니다.
- [미래지향적 패션 어드바이저](https://docs.claude.com/ko/resources/prompt-library/futuristic-fashion-advisor.md): 사용자의 특정 선호도에 맞는 아방가르드 패션 트렌드와 스타일을 제안합니다.
- [Git 마스터하기](https://docs.claude.com/ko/resources/prompt-library/git-gud.md): 사용자가 설명한 버전 관리 작업을 기반으로 적절한 Git 명령어를 생성합니다.
- [Google apps scripter](https://docs.claude.com/ko/resources/prompt-library/google-apps-scripter.md): 사용자 요구사항에 따라 작업을 완료하기 위한 Google Apps 스크립트를 생성합니다.
- [채점 전문가](https://docs.claude.com/ko/resources/prompt-library/grading-guru.md): 사용자 정의 기준과 표준에 따라 작성된 텍스트의 품질을 비교하고 평가합니다.
- [문법 지니](https://docs.claude.com/ko/resources/prompt-library/grammar-genie.md): 문법적으로 틀린 문장을 올바른 영어로 변환합니다.
- [유머러스한 도우미 Hal](https://docs.claude.com/ko/resources/prompt-library/hal-the-humorous-helper.md): 비꼬는 면이 있는 박식한 AI와 채팅하세요.
- [관용구 해설자](https://docs.claude.com/ko/resources/prompt-library/idiom-illuminator.md): 일반적인 관용구와 속담의 의미와 기원을 설명합니다.
- [면접 질문 작성자](https://docs.claude.com/ko/resources/prompt-library/interview-question-crafter.md): 면접을 위한 질문을 생성합니다.
- [LaTeX 전문가](https://docs.claude.com/ko/resources/prompt-library/latex-legend.md): LaTeX 문서를 작성하고 수학 방정식, 표 등을 위한 코드를 생성합니다.
- [수업 계획 작성자](https://docs.claude.com/ko/resources/prompt-library/lesson-planner.md): 모든 주제에 대한 심층적인 수업 계획을 작성합니다.
- [프롬프트 라이브러리](https://docs.claude.com/ko/resources/prompt-library/library.md)
- [마스터 모더레이터](https://docs.claude.com/ko/resources/prompt-library/master-moderator.md): 잠재적으로 유해하거나 불법적인 콘텐츠에 대한 사용자 입력을 평가합니다.
- [회의 서기](https://docs.claude.com/ko/resources/prompt-library/meeting-scribe.md): 회의를 토론 주제, 핵심 요점, 실행 항목을 포함한 간결한 요약으로 정리합니다.
- [메모 마에스트로](https://docs.claude.com/ko/resources/prompt-library/memo-maestro.md): 핵심 포인트를 바탕으로 포괄적인 회사 메모를 작성합니다.
- [마음챙김 멘토](https://docs.claude.com/ko/resources/prompt-library/mindfulness-mentor.md): 사용자가 스트레스 감소를 위한 마음챙김 운동과 기법을 익힐 수 있도록 안내합니다.
- [기분 색상화기](https://docs.claude.com/ko/resources/prompt-library/mood-colorizer.md): 기분에 대한 텍스트 설명을 해당하는 HEX 코드로 변환합니다.
- [동기부여 뮤즈](https://docs.claude.com/ko/resources/prompt-library/motivational-muse.md): 사용자 입력을 바탕으로 개인 맞춤형 동기부여 메시지와 긍정적 확언을 제공합니다.
- [신조어 창조자](https://docs.claude.com/ko/resources/prompt-library/neologism-creator.md): 사용자가 제공한 개념이나 아이디어를 바탕으로 새로운 단어를 만들고 그 정의를 제공합니다.
- [관점 숙고자](https://docs.claude.com/ko/resources/prompt-library/perspectives-ponderer.md): 사용자가 제공한 주제의 장단점을 검토합니다.
- [철학적 사색](https://docs.claude.com/ko/resources/prompt-library/philosophical-musings.md): 깊이 있는 철학적 토론과 사고 실험에 참여하세요.
- [PII 정화기](https://docs.claude.com/ko/resources/prompt-library/pii-purifier.md): 텍스트 문서에서 개인 식별 정보(PII)를 자동으로 감지하고 제거합니다.
- [다국어 초능력](https://docs.claude.com/ko/resources/prompt-library/polyglot-superpowers.md): 모든 언어의 텍스트를 다른 언어로 번역합니다.
- [혼성어 시인](https://docs.claude.com/ko/resources/prompt-library/portmanteau-poet.md): 두 단어를 혼합하여 새롭고 의미 있는 혼성어를 만들어보세요.
- [제품 네이밍 전문가](https://docs.claude.com/ko/resources/prompt-library/product-naming-pro.md): 설명과 키워드로부터 매력적인 제품명을 만들어보세요.
- [산문 다듬기](https://docs.claude.com/ko/resources/prompt-library/prose-polisher.md): 고급 편집 기법과 제안으로 작성된 콘텐츠를 다듬고 개선합니다.
- [말장난 전문가](https://docs.claude.com/ko/resources/prompt-library/pun-dit.md): 주어진 주제를 바탕으로 영리한 말장난과 언어유희를 생성합니다.
- [Python 버그 해결사](https://docs.claude.com/ko/resources/prompt-library/python-bug-buster.md): Python 코드의 버그를 감지하고 수정합니다.
- [리뷰 분류기](https://docs.claude.com/ko/resources/prompt-library/review-classifier.md): 피드백을 미리 지정된 태그와 분류로 분류합니다.
- [수수께끼를 내어보세요](https://docs.claude.com/ko/resources/prompt-library/riddle-me-this.md): 수수께끼를 생성하고 사용자가 해답에 도달할 수 있도록 안내합니다.
- [공상과학 시나리오 시뮬레이터](https://docs.claude.com/ko/resources/prompt-library/sci-fi-scenario-simulator.md): 사용자와 다양한 공상과학 시나리오 및 관련 도전과제와 고려사항에 대해 논의합니다.
- [초등학교 2학년 수준 단순화기](https://docs.claude.com/ko/resources/prompt-library/second-grade-simplifier.md): 복잡한 텍스트를 어린 학습자들이 이해하기 쉽게 만들어 줍니다.
- [직유 전문가](https://docs.claude.com/ko/resources/prompt-library/simile-savant.md): 기본적인 설명으로부터 직유를 생성합니다.
- [소크라테스식 현자](https://docs.claude.com/ko/resources/prompt-library/socratic-sage.md): 사용자가 제공한 주제에 대해 소크라테스식 대화를 나눕니다.
- [스프레드시트 마법사](https://docs.claude.com/ko/resources/prompt-library/spreadsheet-sorcerer.md): 다양한 유형의 데이터로 CSV 스프레드시트를 생성합니다.
- [SQL 마법사](https://docs.claude.com/ko/resources/prompt-library/sql-sorcerer.md): 일상 언어를 SQL 쿼리로 변환합니다.
- [스토리텔링 조수](https://docs.claude.com/ko/resources/prompt-library/storytelling-sidekick.md): 사용자와 협력하여 흥미진진한 이야기를 만들고, 플롯 반전과 캐릭터 개발을 제공합니다.
- [시간 여행 컨설턴트](https://docs.claude.com/ko/resources/prompt-library/time-travel-consultant.md): 사용자가 가상의 시간 여행 시나리오와 그 의미를 탐색할 수 있도록 도와줍니다.
- [잰말놀이](https://docs.claude.com/ko/resources/prompt-library/tongue-twister.md): 도전적인 잰말놀이를 만들어보세요.
- [퀴즈 생성기](https://docs.claude.com/ko/resources/prompt-library/trivia-generator.md): 다양한 주제에 대한 퀴즈 문제를 생성하고 필요할 때 힌트를 제공합니다.
- [트윗 톤 감지기](https://docs.claude.com/ko/resources/prompt-library/tweet-tone-detector.md): 트윗 뒤에 숨겨진 톤과 감정을 감지합니다.
- [VR 피트니스 혁신가](https://docs.claude.com/ko/resources/prompt-library/vr-fitness-innovator.md): 가상현실 피트니스 게임을 위한 창의적인 아이디어를 브레인스토밍하세요.
- [웹사이트 마법사](https://docs.claude.com/ko/resources/prompt-library/website-wizard.md): 사용자 사양에 따라 원페이지 웹사이트를 생성합니다.
- [빠른 시작](https://docs.claude.com/ko/resources/quickstarts.md): Claude와 Claude API로 빠르게 시작하기
- [시스템 상태](https://docs.claude.com/ko/resources/status.md)
- [Obter Chave API](https://docs.claude.com/pt/api/admin-api/apikeys/get-api-key.md)
- [Listar Chaves de API](https://docs.claude.com/pt/api/admin-api/apikeys/list-api-keys.md)
- [Atualizar Chaves de API](https://docs.claude.com/pt/api/admin-api/apikeys/update-api-key.md)
- [Obter Relatório de Uso do Claude Code](https://docs.claude.com/pt/api/admin-api/claude-code/get-claude-code-usage-report.md): Retrieve daily aggregated usage metrics for Claude Code users.
Enables organizations to analyze developer productivity and build custom dashboards.
- [Criar Convite](https://docs.claude.com/pt/api/admin-api/invites/create-invite.md)
- [Excluir Convite](https://docs.claude.com/pt/api/admin-api/invites/delete-invite.md)
- [Obter Convite](https://docs.claude.com/pt/api/admin-api/invites/get-invite.md)
- [Listar Convites](https://docs.claude.com/pt/api/admin-api/invites/list-invites.md)
- [Obter Informações da Organização](https://docs.claude.com/pt/api/admin-api/organization/get-me.md)
- [Obter Relatório de Custos](https://docs.claude.com/pt/api/admin-api/usage-cost/get-cost-report.md)
- [Obter Relatório de Uso para a API de Mensagens](https://docs.claude.com/pt/api/admin-api/usage-cost/get-messages-usage-report.md)
- [Obter Usuário](https://docs.claude.com/pt/api/admin-api/users/get-user.md)
- [Listar Usuários](https://docs.claude.com/pt/api/admin-api/users/list-users.md)
- [Remover Usuário](https://docs.claude.com/pt/api/admin-api/users/remove-user.md)
- [Atualizar Usuário](https://docs.claude.com/pt/api/admin-api/users/update-user.md)
- [Adicionar Membro do Workspace](https://docs.claude.com/pt/api/admin-api/workspace_members/create-workspace-member.md)
- [Excluir Membro do Workspace](https://docs.claude.com/pt/api/admin-api/workspace_members/delete-workspace-member.md)
- [Obter Membro do Workspace](https://docs.claude.com/pt/api/admin-api/workspace_members/get-workspace-member.md)
- [Listar Membros do Espaço de Trabalho](https://docs.claude.com/pt/api/admin-api/workspace_members/list-workspace-members.md)
- [Atualizar Membro do Workspace](https://docs.claude.com/pt/api/admin-api/workspace_members/update-workspace-member.md)
- [Arquivar Workspace](https://docs.claude.com/pt/api/admin-api/workspaces/archive-workspace.md)
- [Criar Workspace](https://docs.claude.com/pt/api/admin-api/workspaces/create-workspace.md)
- [Obter Workspace](https://docs.claude.com/pt/api/admin-api/workspaces/get-workspace.md)
- [Listar Espaços de Trabalho](https://docs.claude.com/pt/api/admin-api/workspaces/list-workspaces.md)
- [Atualizar Workspace](https://docs.claude.com/pt/api/admin-api/workspaces/update-workspace.md)
- [Visão geral da API Admin](https://docs.claude.com/pt/api/administration-api.md)
- [Cabeçalhos beta](https://docs.claude.com/pt/api/beta-headers.md): Documentação para usar cabeçalhos beta com a API Claude
- [Cancelar um Lote de Mensagens](https://docs.claude.com/pt/api/canceling-message-batches.md): Batches may be canceled any time before processing ends. Once cancellation is initiated, the batch enters a `canceling` state, at which time the system may complete any in-progress, non-interruptible requests before finalizing cancellation.
The number of canceled requests is specified in `request_counts`. To determine which requests were canceled, check the individual results within the batch. Note that cancellation may not result in any canceled requests if they were non-interruptible.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [API de Análise de Código Claude](https://docs.claude.com/pt/api/claude-code-analytics-api.md): Acesse programaticamente as análises de uso do Claude Code da sua organização e métricas de produtividade com a API de Administração de Análise de Código Claude.
- [API do Amazon Bedrock](https://docs.claude.com/pt/api/claude-on-amazon-bedrock.md): Os modelos Claude da Anthropic estão agora disponíveis de forma geral através do Amazon Bedrock.
- [API do Vertex AI](https://docs.claude.com/pt/api/claude-on-vertex-ai.md): Os modelos Claude da Anthropic estão agora disponíveis de forma geral através do [Vertex AI](https://cloud.google.com/vertex-ai).
- [SDKs do Cliente](https://docs.claude.com/pt/api/client-sdks.md): Fornecemos bibliotecas cliente em várias linguagens populares que facilitam o trabalho com a API Claude.
- [Criar um Lote de Mensagens](https://docs.claude.com/pt/api/creating-message-batches.md): Send a batch of Message creation requests.
The Message Batches API can be used to process multiple Messages API requests at once. Once a Message Batch is created, it begins processing immediately. Batches can take up to 24 hours to complete.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Excluir um Lote de Mensagens](https://docs.claude.com/pt/api/deleting-message-batches.md): Delete a Message Batch.
Message Batches can only be deleted once they've finished processing. If you'd like to delete an in-progress batch, you must first cancel it.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Erros](https://docs.claude.com/pt/api/errors.md)
- [Baixar um Arquivo](https://docs.claude.com/pt/api/files-content.md): Baixar o conteúdo de um arquivo gerado pelo Claude
- [Criar um Arquivo](https://docs.claude.com/pt/api/files-create.md): Fazer upload de um arquivo
- [Excluir um Arquivo](https://docs.claude.com/pt/api/files-delete.md): Tornar um arquivo inacessível através da API
- [Listar Arquivos](https://docs.claude.com/pt/api/files-list.md): Listar arquivos dentro de um workspace
- [Obter Metadados do Arquivo](https://docs.claude.com/pt/api/files-metadata.md)
- [Obtendo ajuda](https://docs.claude.com/pt/api/getting-help.md): Tentamos fornecer as respostas para as perguntas mais comuns nesta documentação. No entanto, se você precisar de suporte técnico adicional usando Claude, a API Claude, ou qualquer um de nossos produtos, você pode entrar em contato com nossa equipe de suporte em [support.claude.com](https://support.claude.com).
- [Lidando com razões de parada](https://docs.claude.com/pt/api/handling-stop-reasons.md): Entenda como lidar adequadamente com diferentes valores de stop_reason na API Messages do Claude para construir aplicações robustas.
- [Endereços IP](https://docs.claude.com/pt/api/ip-addresses.md): Os serviços da Anthropic usam endereços IP fixos para conexões de entrada e saída. Você pode usar esses endereços para configurar suas regras de firewall para acesso seguro à API Claude e ao Console. Esses endereços não serão alterados sem aviso prévio.
- [Listar Lotes de Mensagens](https://docs.claude.com/pt/api/listing-message-batches.md): List all Message Batches within a Workspace. Most recently created batches are returned first.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Mensagens](https://docs.claude.com/pt/api/messages.md): Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless multi-turn conversations.
Learn more about the Messages API in our [user guide](/en/docs/initial-setup)
- [Exemplos de Lotes de Mensagens](https://docs.claude.com/pt/api/messages-batch-examples.md): Exemplo de uso para a API de Lotes de Mensagens
- [Contar tokens de mensagem](https://docs.claude.com/pt/api/messages-count-tokens.md): Count the number of tokens in a Message.
The Token Count API can be used to count the number of tokens in a Message, including tools, images, and documents, without creating it.
Learn more about token counting in our [user guide](/en/docs/build-with-claude/token-counting)
- [Exemplos de mensagens](https://docs.claude.com/pt/api/messages-examples.md): Exemplos de solicitação e resposta para a API de Mensagens
- [Migrando de Text Completions](https://docs.claude.com/pt/api/migrating-from-text-completions-to-messages.md): Migrando de Text Completions para Messages
- [Obter um Modelo](https://docs.claude.com/pt/api/models.md): Get a specific model.
The Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.
- [Listar Modelos](https://docs.claude.com/pt/api/models-list.md): List available models.
The Models API response can be used to determine which models are available for use in the API. More recently released models are listed first.
- [Compatibilidade com SDK OpenAI](https://docs.claude.com/pt/api/openai-sdk.md): A Anthropic fornece uma camada de compatibilidade que permite usar o SDK OpenAI para testar a API Claude. Com algumas mudanças no código, você pode avaliar rapidamente as capacidades do modelo Anthropic.
- [Visão Geral](https://docs.claude.com/pt/api/overview.md): Visão geral da API do Claude
- [Gerar um prompt](https://docs.claude.com/pt/api/prompt-tools-generate.md): Gerar um prompt bem escrito
- [Melhorar um prompt](https://docs.claude.com/pt/api/prompt-tools-improve.md): Criar um prompt novo e melhorado guiado por feedback
- [Criar template de um prompt](https://docs.claude.com/pt/api/prompt-tools-templatize.md): Criar template de um prompt identificando e extraindo variáveis
- [Limites de taxa](https://docs.claude.com/pt/api/rate-limits.md): Para mitigar o uso indevido e gerenciar a capacidade em nossa API, implementamos limites sobre quanto uma organização pode usar a API Claude.
- [Recuperar Resultados do Lote de Mensagens](https://docs.claude.com/pt/api/retrieving-message-batch-results.md): Streams the results of a Message Batch as a `.jsonl` file.
Each line in the file is a JSON object containing the result of a single request in the Message Batch. Results are not guaranteed to be in the same order as requests. Use the `custom_id` field to match results to requests.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Recuperar um Lote de Mensagens](https://docs.claude.com/pt/api/retrieving-message-batches.md): This endpoint is idempotent and can be used to poll for Message Batch completion. To access the results of a Message Batch, make a request to the `results_url` field in the response.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Níveis de serviço](https://docs.claude.com/pt/api/service-tiers.md): Diferentes níveis de serviço permitem equilibrar disponibilidade, desempenho e custos previsíveis com base nas necessidades da sua aplicação.
- [Regiões suportadas](https://docs.claude.com/pt/api/supported-regions.md): Aqui estão os países, regiões e territórios dos quais podemos atualmente suportar acesso:
- [API de Uso e Custo](https://docs.claude.com/pt/api/usage-cost-api.md): Acesse programaticamente os dados de uso e custo da API da sua organização com a API de Administração de Uso e Custo.
- [Versões](https://docs.claude.com/pt/api/versioning.md): Ao fazer solicitações de API, você deve enviar um cabeçalho de solicitação `anthropic-version`. Por exemplo, `anthropic-version: 2023-06-01`. Se você estiver usando nossos [SDKs de cliente](/pt/api/client-sdks), isso é tratado automaticamente para você.
- [Glossário](https://docs.claude.com/pt/docs/about-claude/glossary.md): Estes conceitos não são únicos aos modelos de linguagem da Anthropic, mas apresentamos um breve resumo dos termos-chave abaixo.
- [Depreciações de modelos](https://docs.claude.com/pt/docs/about-claude/model-deprecations.md)
- [Escolhendo o modelo certo](https://docs.claude.com/pt/docs/about-claude/models/choosing-a-model.md): Selecionar o modelo Claude ideal para sua aplicação envolve equilibrar três considerações principais: capacidades, velocidade e custo. Este guia ajuda você a tomar uma decisão informada com base em seus requisitos específicos.
- [Migrando para o Claude 4](https://docs.claude.com/pt/docs/about-claude/models/migrating-to-claude-4.md): Guia para migrar dos modelos Claude 3.7 para os modelos Claude 4 (Opus 4 e Sonnet 4).
- [Visão geral dos modelos](https://docs.claude.com/pt/docs/about-claude/models/overview.md): Claude é uma família de modelos de linguagem grandes de última geração desenvolvidos pela Anthropic. Este guia apresenta nossos modelos e compara seu desempenho com modelos legados.
- [Preços](https://docs.claude.com/pt/docs/about-claude/pricing.md): Aprenda sobre a estrutura de preços da Anthropic para modelos e recursos
- [Moderação de conteúdo](https://docs.claude.com/pt/docs/about-claude/use-case-guides/content-moderation.md): A moderação de conteúdo é um aspecto crítico para manter um ambiente seguro, respeitoso e produtivo em aplicações digitais. Neste guia, discutiremos como Claude pode ser usado para moderar conteúdo dentro da sua aplicação digital.
- [Agente de suporte ao cliente](https://docs.claude.com/pt/docs/about-claude/use-case-guides/customer-support-chat.md): Este guia explica como aproveitar as capacidades conversacionais avançadas do Claude para lidar com consultas de clientes em tempo real, fornecendo suporte 24/7, reduzindo tempos de espera e gerenciando altos volumes de suporte com respostas precisas e interações positivas.
- [Resumo jurídico](https://docs.claude.com/pt/docs/about-claude/use-case-guides/legal-summarization.md): Este guia explica como aproveitar as capacidades avançadas de processamento de linguagem natural do Claude para resumir eficientemente documentos jurídicos, extraindo informações-chave e agilizando a pesquisa jurídica. Com o Claude, você pode otimizar a revisão de contratos, preparação de litígios e trabalho regulatório, economizando tempo e garantindo precisão em seus processos jurídicos.
- [Guias para casos de uso comuns](https://docs.claude.com/pt/docs/about-claude/use-case-guides/overview.md)
- [Roteamento de tickets](https://docs.claude.com/pt/docs/about-claude/use-case-guides/ticket-routing.md): Este guia explica como aproveitar as capacidades avançadas de compreensão de linguagem natural do Claude para classificar tickets de suporte ao cliente em escala com base na intenção do cliente, urgência, priorização, perfil do cliente e muito mais.
- [Complemento do Google Sheets](https://docs.claude.com/pt/docs/agents-and-tools/claude-for-sheets.md): A [extensão Claude for Sheets](https://workspace.google.com/marketplace/app/claude%5Ffor%5Fsheets/909417792257) integra o Claude ao Google Sheets, permitindo que você execute interações com o Claude diretamente nas células.
- [Conector MCP](https://docs.claude.com/pt/docs/agents-and-tools/mcp-connector.md): O recurso de conector do Protocolo de Contexto de Modelo (MCP) do Claude permite conectar-se a servidores MCP remotos diretamente da API de Mensagens sem um cliente MCP separado.
- [Servidores MCP remotos](https://docs.claude.com/pt/docs/agents-and-tools/remote-mcp-servers.md): Servidores MCP remotos
- [Ferramenta bash](https://docs.claude.com/pt/docs/agents-and-tools/tool-use/bash-tool.md): A ferramenta bash permite que Claude execute comandos shell em uma sessão bash persistente, permitindo operações do sistema, execução de scripts e automação de linha de comando.
- [Ferramenta de execução de código](https://docs.claude.com/pt/docs/agents-and-tools/tool-use/code-execution-tool.md)
- [Ferramenta de uso do computador](https://docs.claude.com/pt/docs/agents-and-tools/tool-use/computer-use-tool.md): Claude pode interagir com ambientes de computador através da ferramenta de uso do computador, que fornece capacidades de captura de tela e controle de mouse/teclado para interação autônoma com desktop.
- [Streaming de ferramentas granular](https://docs.claude.com/pt/docs/agents-and-tools/tool-use/fine-grained-tool-streaming.md): O uso de ferramentas agora suporta streaming granular para valores de parâmetros, permitindo que os desenvolvedores façam streaming de parâmetros de uso de ferramentas sem buffering/validação JSON.
- [Como implementar o uso de ferramentas](https://docs.claude.com/pt/docs/agents-and-tools/tool-use/implement-tool-use.md): Aprenda como implementar o uso de ferramentas com Claude, incluindo definições de ferramentas, controle de saída e melhores práticas.
- [Uso de ferramentas com Claude](https://docs.claude.com/pt/docs/agents-and-tools/tool-use/overview.md)
- [Ferramenta de editor de texto](https://docs.claude.com/pt/docs/agents-and-tools/tool-use/text-editor-tool.md): Claude pode usar uma ferramenta de editor de texto definida pela Anthropic para visualizar e modificar arquivos de texto, ajudando você a depurar, corrigir e melhorar seu código ou outros documentos de texto.
- [Uso eficiente de tokens com ferramentas](https://docs.claude.com/pt/docs/agents-and-tools/tool-use/token-efficient-tool-use.md)
- [Ferramenta de busca web](https://docs.claude.com/pt/docs/agents-and-tools/tool-use/web-fetch-tool.md): A ferramenta de busca web permite que Claude recupere conteúdo completo de páginas web e documentos PDF especificados.
- [Ferramenta de busca na web](https://docs.claude.com/pt/docs/agents-and-tools/tool-use/web-search-tool.md): A ferramenta de busca na web dá ao Claude acesso direto ao conteúdo web em tempo real, permitindo que responda perguntas com informações atualizadas além de seu limite de conhecimento.
- [Processamento em lote](https://docs.claude.com/pt/docs/build-with-claude/batch-processing.md)
- [Citações](https://docs.claude.com/pt/docs/build-with-claude/citations.md)
- [Janelas de contexto](https://docs.claude.com/pt/docs/build-with-claude/context-windows.md)
- [Embeddings](https://docs.claude.com/pt/docs/build-with-claude/embeddings.md): Embeddings de texto são representações numéricas de texto que permitem medir similaridade semântica. Este guia apresenta embeddings, suas aplicações e como usar modelos de embedding para tarefas como busca, recomendações e detecção de anomalias.
- [Construindo com pensamento estendido](https://docs.claude.com/pt/docs/build-with-claude/extended-thinking.md)
- [API de Arquivos](https://docs.claude.com/pt/docs/build-with-claude/files.md): A API de Arquivos permite fazer upload e gerenciar arquivos para usar com a API Claude sem precisar reenviar conteúdo a cada solicitação.
- [Suporte multilíngue](https://docs.claude.com/pt/docs/build-with-claude/multilingual-support.md): Claude se destaca em tarefas em vários idiomas, mantendo um forte desempenho interlinguístico em relação ao inglês.
- [Visão geral dos recursos](https://docs.claude.com/pt/docs/build-with-claude/overview.md): Explore os recursos e capacidades avançados do Claude.
- [Suporte a PDF](https://docs.claude.com/pt/docs/build-with-claude/pdf-support.md): Processe PDFs com Claude. Extraia texto, analise gráficos e compreenda conteúdo visual de seus documentos.
- [Cache de prompt](https://docs.claude.com/pt/docs/build-with-claude/prompt-caching.md)
- [Seja claro, direto e detalhado](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/be-clear-and-direct.md)
- [Deixe o Claude pensar (prompting de cadeia de pensamento) para aumentar o desempenho](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/chain-of-thought.md)
- [Encadeie prompts complexos para um desempenho mais forte](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/chain-prompts.md)
- [Melhores práticas de engenharia de prompt do Claude 4](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/claude-4-best-practices.md): Guia específico de técnicas de engenharia de prompt para modelos Claude 4 (Opus 4.1, Opus 4 e Sonnet 4) para ajudá-lo a obter resultados ideais em suas aplicações.
- [Dicas de pensamento estendido](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/extended-thinking-tips.md)
- [Dicas para prompts com contexto longo](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/long-context-tips.md)
- [Use exemplos (prompting multishot) para guiar o comportamento do Claude](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/multishot-prompting.md)
- [Visão geral da engenharia de prompts](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/overview.md)
- [Preencher previamente a resposta do Claude para maior controle de saída](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/prefill-claudes-response.md)
- [Gerar automaticamente modelos de prompt de primeiro rascunho](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/prompt-generator.md)
- [Use nosso melhorador de prompts para otimizar seus prompts](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/prompt-improver.md): Use nosso melhorador de prompts para otimizar seus prompts
- [Use modelos de prompt e variáveis](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/prompt-templates-and-variables.md): Aprenda a usar modelos de prompt e variáveis para criar aplicações mais eficientes e escaláveis com Claude.
- [Atribuindo um papel ao Claude com um prompt de sistema](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/system-prompts.md)
- [Use tags XML para estruturar seus prompts](https://docs.claude.com/pt/docs/build-with-claude/prompt-engineering/use-xml-tags.md)
- [Resultados de pesquisa](https://docs.claude.com/pt/docs/build-with-claude/search-results.md): Habilite citações naturais para aplicações RAG fornecendo resultados de pesquisa com atribuição de fonte
- [Streaming de Mensagens](https://docs.claude.com/pt/docs/build-with-claude/streaming.md): Aprenda como usar streaming para receber respostas incrementais usando server-sent events (SSE) ao criar mensagens.
- [Contagem de tokens](https://docs.claude.com/pt/docs/build-with-claude/token-counting.md): A contagem de tokens permite determinar o número de tokens em uma mensagem antes de enviá-la para Claude, ajudando você a tomar decisões informadas sobre seus prompts e uso.
- [Visão](https://docs.claude.com/pt/docs/build-with-claude/vision.md): As famílias de modelos Claude 3 e 4 vêm com novas capacidades de visão que permitem ao Claude entender e analisar imagens, abrindo possibilidades empolgantes para interação multimodal.
- [Claude Code no Amazon Bedrock](https://docs.claude.com/pt/docs/claude-code/amazon-bedrock.md): Aprenda sobre como configurar o Claude Code através do Amazon Bedrock, incluindo configuração, configuração do IAM e solução de problemas.
- [Análises](https://docs.claude.com/pt/docs/claude-code/analytics.md): Visualize insights detalhados de uso e métricas de produtividade para a implantação do Claude Code da sua organização.
- [Referência da CLI](https://docs.claude.com/pt/docs/claude-code/cli-reference.md): Referência completa para a interface de linha de comando do Claude Code, incluindo comandos e flags.
- [Fluxos de trabalho comuns](https://docs.claude.com/pt/docs/claude-code/common-workflows.md): Aprenda sobre fluxos de trabalho comuns com Claude Code.
- [Gerir custos de forma eficaz](https://docs.claude.com/pt/docs/claude-code/costs.md): Aprenda como rastrear e otimizar o uso de tokens e custos ao usar Claude Code.
- [Uso de dados](https://docs.claude.com/pt/docs/claude-code/data-usage.md): Aprenda sobre as políticas de uso de dados da Anthropic para Claude
- [Contêineres de desenvolvimento](https://docs.claude.com/pt/docs/claude-code/devcontainer.md): Aprenda sobre o contêiner de desenvolvimento Claude Code para equipes que precisam de ambientes consistentes e seguros.
- [Claude Code GitHub Actions](https://docs.claude.com/pt/docs/claude-code/github-actions.md): Aprenda sobre a integração do Claude Code no seu fluxo de trabalho de desenvolvimento com Claude Code GitHub Actions
- [Claude Code GitLab CI/CD](https://docs.claude.com/pt/docs/claude-code/gitlab-ci-cd.md): Aprenda sobre a integração do Claude Code no seu fluxo de trabalho de desenvolvimento com GitLab CI/CD
- [Claude Code no Google Vertex AI](https://docs.claude.com/pt/docs/claude-code/google-vertex-ai.md): Aprenda sobre configurar o Claude Code através do Google Vertex AI, incluindo configuração, configuração IAM e solução de problemas.
- [Referência de hooks](https://docs.claude.com/pt/docs/claude-code/hooks.md): Esta página fornece documentação de referência para implementar hooks no Claude Code.
- [Comece com os hooks do Claude Code](https://docs.claude.com/pt/docs/claude-code/hooks-guide.md): Aprenda como personalizar e estender o comportamento do Claude Code registrando comandos shell
- [Gerenciamento de Identidade e Acesso](https://docs.claude.com/pt/docs/claude-code/iam.md): Aprenda como configurar autenticação de usuário, autorização e controles de acesso para Claude Code em sua organização.
- [Adicionar Claude Code ao seu IDE](https://docs.claude.com/pt/docs/claude-code/ide-integrations.md): Aprenda como adicionar Claude Code ao seu IDE favorito
- [Modo interativo](https://docs.claude.com/pt/docs/claude-code/interactive-mode.md): Referência completa para atalhos de teclado, modos de entrada e recursos interativos em sessões do Claude Code.
- [Legal e conformidade](https://docs.claude.com/pt/docs/claude-code/legal-and-compliance.md): Acordos legais, certificações de conformidade e informações de segurança para Claude Code.
- [Configuração de gateway LLM](https://docs.claude.com/pt/docs/claude-code/llm-gateway.md): Aprenda como configurar o Claude Code com soluções de gateway LLM, incluindo configuração do LiteLLM, métodos de autenticação e recursos empresariais como rastreamento de uso e gerenciamento de orçamento.
- [Conectar Claude Code a ferramentas via MCP](https://docs.claude.com/pt/docs/claude-code/mcp.md): Aprenda como conectar Claude Code às suas ferramentas com o Model Context Protocol.
- [Gerenciar a memória do Claude](https://docs.claude.com/pt/docs/claude-code/memory.md): Aprenda como gerenciar a memória do Claude Code entre sessões com diferentes localizações de memória e melhores práticas.
- [Configuração do modelo](https://docs.claude.com/pt/docs/claude-code/model-config.md): Aprenda sobre a configuração do modelo Claude Code, incluindo aliases de modelo como `opusplan`
- [Monitoramento](https://docs.claude.com/pt/docs/claude-code/monitoring-usage.md): Aprenda como habilitar e configurar OpenTelemetry para Claude Code.
- [Configuração de rede empresarial](https://docs.claude.com/pt/docs/claude-code/network-config.md): Configure o Claude Code para ambientes empresariais com servidores proxy, Autoridades de Certificação (CA) personalizadas e autenticação Transport Layer Security (mTLS) mútua.
- [Estilos de saída](https://docs.claude.com/pt/docs/claude-code/output-styles.md): Adapte o Claude Code para usos além da engenharia de software
- [Visão geral do Claude Code](https://docs.claude.com/pt/docs/claude-code/overview.md): Aprenda sobre o Claude Code, a ferramenta de codificação agêntica da Anthropic que vive no seu terminal e ajuda você a transformar ideias em código mais rápido do que nunca.
- [Início Rápido](https://docs.claude.com/pt/docs/claude-code/quickstart.md): Bem-vindo ao Claude Code!
- [Ferramentas Personalizadas](https://docs.claude.com/pt/docs/claude-code/sdk/custom-tools.md): Construa e integre ferramentas personalizadas para estender a funcionalidade do Claude Code SDK
- [Modificando prompts do sistema](https://docs.claude.com/pt/docs/claude-code/sdk/modifying-system-prompts.md): Aprenda como personalizar o comportamento do Claude modificando prompts do sistema usando três abordagens - estilos de saída, appendSystemPrompt e customSystemPrompt.
- [Rastreamento de Custos e Uso](https://docs.claude.com/pt/docs/claude-code/sdk/sdk-cost-tracking.md): Entenda e rastreie o uso de tokens para faturamento no Claude Code SDK
- [Modo headless](https://docs.claude.com/pt/docs/claude-code/sdk/sdk-headless.md): Execute Claude Code programaticamente sem interface de usuário interativa
- [MCP no SDK](https://docs.claude.com/pt/docs/claude-code/sdk/sdk-mcp.md): Estenda o Claude Code com ferramentas personalizadas usando servidores do Model Context Protocol
- [Visão Geral](https://docs.claude.com/pt/docs/claude-code/sdk/sdk-overview.md): Construa agentes de IA personalizados com o Claude Code SDK
- [Gerenciamento de Permissões](https://docs.claude.com/pt/docs/claude-code/sdk/sdk-permissions.md): Controle o uso de ferramentas e permissões no Claude Code SDK
- [Referência do SDK Python](https://docs.claude.com/pt/docs/claude-code/sdk/sdk-python.md): Referência completa da API para o SDK Python do Claude Code, incluindo todas as funções, tipos e classes.
- [Gerenciamento de Sessões](https://docs.claude.com/pt/docs/claude-code/sdk/sdk-sessions.md): Entendendo como o Claude Code SDK lida com sessões e retomada de sessões
- [Comandos Slash no SDK](https://docs.claude.com/pt/docs/claude-code/sdk/sdk-slash-commands.md): Aprenda como usar comandos slash para controlar sessões do Claude Code através do SDK
- [Referência do SDK TypeScript](https://docs.claude.com/pt/docs/claude-code/sdk/sdk-typescript.md): Referência completa da API para o SDK TypeScript do Claude Code, incluindo todas as funções, tipos e interfaces.
- [Entrada de Streaming](https://docs.claude.com/pt/docs/claude-code/sdk/streaming-vs-single-mode.md): Compreendendo os dois modos de entrada para o Claude Code SDK e quando usar cada um
- [Subagentes no SDK](https://docs.claude.com/pt/docs/claude-code/sdk/subagents.md): Trabalhando com subagentes no Claude Code SDK
- [Listas de Tarefas](https://docs.claude.com/pt/docs/claude-code/sdk/todo-tracking.md): Rastreie e exiba tarefas usando o Claude Code SDK para gerenciamento organizado de tarefas
- [Segurança](https://docs.claude.com/pt/docs/claude-code/security.md): Aprenda sobre as proteções de segurança do Claude Code e as melhores práticas para uso seguro.
- [Configurações do Claude Code](https://docs.claude.com/pt/docs/claude-code/settings.md): Configure o Claude Code com configurações globais e de nível de projeto, e variáveis de ambiente.
- [Configurar Claude Code](https://docs.claude.com/pt/docs/claude-code/setup.md): Instale, autentique e comece a usar Claude Code em sua máquina de desenvolvimento.
- [Comandos slash](https://docs.claude.com/pt/docs/claude-code/slash-commands.md): Controle o comportamento do Claude durante uma sessão interativa com comandos slash.
- [Configuração da linha de status](https://docs.claude.com/pt/docs/claude-code/statusline.md): Crie uma linha de status personalizada para o Claude Code para exibir informações contextuais
- [Subagentes](https://docs.claude.com/pt/docs/claude-code/sub-agents.md): Crie e use subagentes de IA especializados no Claude Code para fluxos de trabalho específicos de tarefas e gerenciamento de contexto aprimorado.
- [Otimize sua configuração de terminal](https://docs.claude.com/pt/docs/claude-code/terminal-config.md): Claude Code funciona melhor quando seu terminal está configurado adequadamente. Siga estas diretrizes para otimizar sua experiência.
- [Visão geral da implantação empresarial](https://docs.claude.com/pt/docs/claude-code/third-party-integrations.md): Aprenda como o Claude Code pode se integrar com vários serviços de terceiros e infraestrutura para atender aos requisitos de implantação empresarial.
- [Solução de problemas](https://docs.claude.com/pt/docs/claude-code/troubleshooting.md): Descubra soluções para problemas comuns com a instalação e uso do Claude Code.
- [Comece com Claude](https://docs.claude.com/pt/docs/get-started.md): Faça sua primeira chamada de API para Claude e construa um assistente simples de busca na web
- [Introdução ao Claude](https://docs.claude.com/pt/docs/intro.md): Claude é uma plataforma de IA altamente performática, confiável e inteligente construída pela Anthropic. Claude se destaca em tarefas envolvendo linguagem, raciocínio, análise, codificação e muito mais.
- [Política de Privacidade da Anthropic](https://docs.claude.com/pt/docs/legal-center/privacy.md)
- [Segurança e conformidade](https://docs.claude.com/pt/docs/legal-center/security-compliance.md)
- [Construindo com Claude](https://docs.claude.com/pt/docs/overview.md): Claude é uma família de [modelos de IA altamente performáticos e inteligentes](/pt/docs/about-claude/models) construídos pela Anthropic. Embora Claude seja poderoso e extensível, também é a IA mais confiável e confiável disponível. Segue protocolos críticos, comete menos erros e é resistente a jailbreaks—permitindo que [clientes empresariais](https://claude.com/customers) construam as aplicações alimentadas por IA mais seguras em escala.
- [Defina seus critérios de sucesso](https://docs.claude.com/pt/docs/test-and-evaluate/define-success.md)
- [Criar avaliações empíricas robustas](https://docs.claude.com/pt/docs/test-and-evaluate/develop-tests.md): Desenvolver casos de teste
- [Usando a Ferramenta de Avaliação](https://docs.claude.com/pt/docs/test-and-evaluate/eval-tool.md): O [Console Claude](https://console.anthropic.com/dashboard) apresenta uma **ferramenta de Avaliação** que permite testar seus prompts sob vários cenários.
- [null](https://docs.claude.com/pt/docs/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals.md)
- [Aumentar a consistência da saída (modo JSON)](https://docs.claude.com/pt/docs/test-and-evaluate/strengthen-guardrails/increase-consistency.md)
- [Mantenha o Claude no personagem com prompts de papel e pré-preenchimento](https://docs.claude.com/pt/docs/test-and-evaluate/strengthen-guardrails/keep-claude-in-character.md)
- [Mitigar jailbreaks e injeções de prompt](https://docs.claude.com/pt/docs/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks.md)
- [Reduzir alucinações](https://docs.claude.com/pt/docs/test-and-evaluate/strengthen-guardrails/reduce-hallucinations.md)
- [Reduzindo a latência](https://docs.claude.com/pt/docs/test-and-evaluate/strengthen-guardrails/reduce-latency.md)
- [Reduzir vazamento de prompt](https://docs.claude.com/pt/docs/test-and-evaluate/strengthen-guardrails/reduce-prompt-leak.md)
- [null](https://docs.claude.com/pt/home.md)
- [Plataforma de Desenvolvedores Claude](https://docs.claude.com/pt/release-notes/api.md): Acompanhe as atualizações da API Claude e do Console Claude.
- [Aplicações Claude](https://docs.claude.com/pt/release-notes/claude-apps.md): Acompanhe as atualizações das aplicações Claude da Anthropic.
- [Claude Code](https://docs.claude.com/pt/release-notes/claude-code.md): Mantenha-se atualizado com os recursos mais recentes e melhorias do Claude Code
- [Visão Geral](https://docs.claude.com/pt/release-notes/overview.md): Acompanhe as atualizações dos produtos e serviços da Anthropic.
- [Prompts do Sistema](https://docs.claude.com/pt/release-notes/system-prompts.md): Veja atualizações dos prompts principais do sistema no [Claude.ai](https://www.claude.ai) e nos aplicativos Claude para [iOS](http://anthropic.com/ios) e [Android](http://anthropic.com/android).
- [Cartão do sistema Claude 3.7](https://docs.claude.com/pt/resources/claude-3-7-system-card.md)
- [Cartão do modelo Claude 3](https://docs.claude.com/pt/resources/claude-3-model-card.md)
- [Cartão de Sistema Claude 4](https://docs.claude.com/pt/resources/claude-4-system-card.md): Cartão de modelo da Anthropic para Claude Opus e Sonnet 4 com detalhes de desempenho e segurança.
- [Cursos da Anthropic](https://docs.claude.com/pt/resources/courses.md)
- [null](https://docs.claude.com/pt/resources/overview.md)
- [Editor adaptativo](https://docs.claude.com/pt/resources/prompt-library/adaptive-editor.md): Reescreva texto seguindo instruções fornecidas pelo usuário, como com um tom, público ou estilo diferente.
- [Analisador de códigos de aeroporto](https://docs.claude.com/pt/resources/prompt-library/airport-code-analyst.md): Encontre e extraia códigos de aeroporto do texto.
- [Antropólogo alienígena](https://docs.claude.com/pt/resources/prompt-library/alien-anthropologist.md): Analise a cultura e costumes humanos da perspectiva de um antropólogo alienígena.
- [Alquimista da aliteração](https://docs.claude.com/pt/resources/prompt-library/alliteration-alchemist.md): Gere frases e sentenças aliterativas para qualquer assunto dado.
- [Transmissões da Babel](https://docs.claude.com/pt/resources/prompt-library/babels-broadcasts.md): Crie tweets cativantes de anúncio de produto nos 10 idiomas mais falados do mundo.
- [Construtor de marca](https://docs.claude.com/pt/resources/prompt-library/brand-builder.md): Crie um briefing de design para uma identidade de marca holística.
- [Coach de carreira](https://docs.claude.com/pt/resources/prompt-library/career-coach.md): Participe de conversas de role-play com um coach de carreira de IA.
- [Cite suas fontes](https://docs.claude.com/pt/resources/prompt-library/cite-your-sources.md): Obtenha respostas a perguntas sobre o conteúdo de um documento com citações relevantes que apoiam a resposta.
- [Esclarecedor de código](https://docs.claude.com/pt/resources/prompt-library/code-clarifier.md): Simplifique e explique código complexo em linguagem simples.
- [Consultor de código](https://docs.claude.com/pt/resources/prompt-library/code-consultant.md): Sugira melhorias para otimizar o desempenho do código Python.
- [Clarividente corporativo](https://docs.claude.com/pt/resources/prompt-library/corporate-clairvoyant.md): Extraia insights, identifique riscos e destile informações-chave de relatórios corporativos longos em um único memorando.
- [Teclas Cósmicas](https://docs.claude.com/pt/resources/prompt-library/cosmic-keystrokes.md): Gere um jogo interativo de digitação rápida em um único arquivo HTML, apresentando jogabilidade de rolagem lateral e estilização Tailwind CSS.
- [Conversor CSV](https://docs.claude.com/pt/resources/prompt-library/csv-converter.md): Converta dados de vários formatos (JSON, XML, etc.) em arquivos CSV devidamente formatados.
- [Criador culinário](https://docs.claude.com/pt/resources/prompt-library/culinary-creator.md): Sugira ideias de receitas baseadas nos ingredientes disponíveis do usuário e preferências dietéticas.
- [Organizador de dados](https://docs.claude.com/pt/resources/prompt-library/data-organizer.md): Transforme texto não estruturado em tabelas JSON personalizadas.
- [Decodificador de direções](https://docs.claude.com/pt/resources/prompt-library/direction-decoder.md): Transforme linguagem natural em instruções passo a passo.
- [Intérprete de sonhos](https://docs.claude.com/pt/resources/prompt-library/dream-interpreter.md): Ofereça interpretações e insights sobre o simbolismo dos sonhos do usuário.
- [Estimador de eficiência](https://docs.claude.com/pt/resources/prompt-library/efficiency-estimator.md): Calcule a complexidade de tempo de funções e algoritmos.
- [Extrator de email](https://docs.claude.com/pt/resources/prompt-library/email-extractor.md): Extrair endereços de email de um documento em uma lista formatada em JSON.
- [Codificador de emoji](https://docs.claude.com/pt/resources/prompt-library/emoji-encoder.md): Converta texto simples em mensagens divertidas e expressivas com emojis.
- [Navegador de dilemas éticos](https://docs.claude.com/pt/resources/prompt-library/ethical-dilemma-navigator.md): Ajude o usuário a pensar através de dilemas éticos complexos e forneça diferentes perspectivas.
- [Especialista em fórmulas do Excel](https://docs.claude.com/pt/resources/prompt-library/excel-formula-expert.md): Crie fórmulas do Excel baseadas em cálculos ou manipulações de dados descritos pelo usuário.
- [Fabricador de funções](https://docs.claude.com/pt/resources/prompt-library/function-fabricator.md): Crie funções Python baseadas em especificações detalhadas.
- [Consultor de moda futurística](https://docs.claude.com/pt/resources/prompt-library/futuristic-fashion-advisor.md): Sugira tendências e estilos de moda vanguardistas para as preferências específicas do usuário.
- [Git gud](https://docs.claude.com/pt/resources/prompt-library/git-gud.md): Gerar comandos Git apropriados baseados em ações de controle de versão descritas pelo usuário.
- [Scripter de aplicativos Google](https://docs.claude.com/pt/resources/prompt-library/google-apps-scripter.md): Gere scripts do Google Apps para completar tarefas baseadas nos requisitos do usuário.
- [Guru de avaliação](https://docs.claude.com/pt/resources/prompt-library/grading-guru.md): Compare e avalie a qualidade de textos escritos com base em critérios e padrões definidos pelo usuário.
- [Gênio da gramática](https://docs.claude.com/pt/resources/prompt-library/grammar-genie.md): Transforme frases gramaticalmente incorretas em inglês adequado.
- [Hal, o assistente humorístico](https://docs.claude.com/pt/resources/prompt-library/hal-the-humorous-helper.md): Converse com uma IA conhecedora que tem um lado sarcástico.
- [Iluminador de expressões idiomáticas](https://docs.claude.com/pt/resources/prompt-library/idiom-illuminator.md): Explique o significado e a origem de expressões idiomáticas e provérbios comuns.
- [Criador de perguntas de entrevista](https://docs.claude.com/pt/resources/prompt-library/interview-question-crafter.md): Gere perguntas para entrevistas.
- [Lenda do LaTeX](https://docs.claude.com/pt/resources/prompt-library/latex-legend.md): Escreva documentos LaTeX, gerando código para equações matemáticas, tabelas e muito mais.
- [Planejador de aulas](https://docs.claude.com/pt/resources/prompt-library/lesson-planner.md): Crie planos de aula detalhados sobre qualquer assunto.
- [Biblioteca de Prompts](https://docs.claude.com/pt/resources/prompt-library/library.md)
- [Moderador mestre](https://docs.claude.com/pt/resources/prompt-library/master-moderator.md): Avalie entradas de usuários para conteúdo potencialmente prejudicial ou ilegal.
- [Secretário de reunião](https://docs.claude.com/pt/resources/prompt-library/meeting-scribe.md): Destile reuniões em resumos concisos incluindo tópicos de discussão, principais conclusões e itens de ação.
- [Memo maestro](https://docs.claude.com/pt/resources/prompt-library/memo-maestro.md): Componha memorandos abrangentes da empresa com base em pontos-chave.
- [Mentor de mindfulness](https://docs.claude.com/pt/resources/prompt-library/mindfulness-mentor.md): Guie o usuário através de exercícios de mindfulness e técnicas para redução do estresse.
- [Colorizador de humor](https://docs.claude.com/pt/resources/prompt-library/mood-colorizer.md): Transforme descrições textuais de humores em códigos HEX correspondentes.
- [Musa motivacional](https://docs.claude.com/pt/resources/prompt-library/motivational-muse.md): Forneça mensagens motivacionais personalizadas e afirmações baseadas na entrada do usuário.
- [Criador de neologismos](https://docs.claude.com/pt/resources/prompt-library/neologism-creator.md): Invente novas palavras e forneça suas definições baseadas em conceitos ou ideias fornecidas pelo usuário.
- [Ponderador de perspectivas](https://docs.claude.com/pt/resources/prompt-library/perspectives-ponderer.md): Avalie os prós e contras de um tópico fornecido pelo usuário.
- [Reflexões filosóficas](https://docs.claude.com/pt/resources/prompt-library/philosophical-musings.md): Envolva-se em discussões filosóficas profundas e experimentos mentais.
- [Purificador de PII](https://docs.claude.com/pt/resources/prompt-library/pii-purifier.md): Detecta e remove automaticamente informações pessoais identificáveis (PII) de documentos de texto.
- [Superpoderes poliglotas](https://docs.claude.com/pt/resources/prompt-library/polyglot-superpowers.md): Traduza texto de qualquer idioma para qualquer idioma.
- [Poeta de palavras-valise](https://docs.claude.com/pt/resources/prompt-library/portmanteau-poet.md): Misture duas palavras para criar uma nova palavra-valise significativa.
- [Especialista em nomes de produtos](https://docs.claude.com/pt/resources/prompt-library/product-naming-pro.md): Crie nomes cativantes de produtos a partir de descrições e palavras-chave.
- [Polidor de prosa](https://docs.claude.com/pt/resources/prompt-library/prose-polisher.md): Refine e melhore conteúdo escrito com técnicas avançadas de edição de texto e sugestões.
- [Especialista em Trocadilhos](https://docs.claude.com/pt/resources/prompt-library/pun-dit.md): Gere trocadilhos inteligentes e jogos de palavras baseados em qualquer tópico dado.
- [Corretor de bugs Python](https://docs.claude.com/pt/resources/prompt-library/python-bug-buster.md): Detecta e corrige bugs em código Python.
- [Classificador de avaliações](https://docs.claude.com/pt/resources/prompt-library/review-classifier.md): Categorize feedback em tags e categorizações pré-especificadas.
- [Decifre-me isso](https://docs.claude.com/pt/resources/prompt-library/riddle-me-this.md): Gere charadas e guie o usuário para as soluções.
- [Simulador de cenários de ficção científica](https://docs.claude.com/pt/resources/prompt-library/sci-fi-scenario-simulator.md): Discuta com o usuário vários cenários de ficção científica e desafios e considerações associados.
- [Simplificador de segundo ano](https://docs.claude.com/pt/resources/prompt-library/second-grade-simplifier.md): Torne textos complexos fáceis de entender para jovens aprendizes.
- [Especialista em símiles](https://docs.claude.com/pt/resources/prompt-library/simile-savant.md): Gere símiles a partir de descrições básicas.
- [Sábio socrático](https://docs.claude.com/pt/resources/prompt-library/socratic-sage.md): Envolva-se em conversas no estilo socrático sobre um tópico fornecido pelo usuário.
- [Feiticeiro de planilhas](https://docs.claude.com/pt/resources/prompt-library/spreadsheet-sorcerer.md): Gere planilhas CSV com vários tipos de dados.
- [Feiticeiro SQL](https://docs.claude.com/pt/resources/prompt-library/sql-sorcerer.md): Transforme linguagem cotidiana em consultas SQL.
- [Companheiro de narrativa](https://docs.claude.com/pt/resources/prompt-library/storytelling-sidekick.md): Crie colaborativamente histórias envolventes com o usuário, oferecendo reviravoltas na trama e desenvolvimento de personagens.
- [Consultor de viagem no tempo](https://docs.claude.com/pt/resources/prompt-library/time-travel-consultant.md): Ajude o usuário a navegar por cenários hipotéticos de viagem no tempo e suas implicações.
- [Trava-línguas](https://docs.claude.com/pt/resources/prompt-library/tongue-twister.md): Crie trava-línguas desafiadores.
- [Gerador de perguntas de trivia](https://docs.claude.com/pt/resources/prompt-library/trivia-generator.md): Gere perguntas de trivia sobre uma ampla gama de tópicos e forneça dicas quando necessário.
- [Detector de tom de tweet](https://docs.claude.com/pt/resources/prompt-library/tweet-tone-detector.md): Detecte o tom e sentimento por trás dos tweets.
- [Inovador de fitness em VR](https://docs.claude.com/pt/resources/prompt-library/vr-fitness-innovator.md): Faça brainstorming de ideias criativas para jogos de fitness em realidade virtual.
- [Assistente de websites](https://docs.claude.com/pt/resources/prompt-library/website-wizard.md): Crie websites de uma página baseados nas especificações do usuário.
- [Guias de Início Rápido](https://docs.claude.com/pt/resources/quickstarts.md): Comece rapidamente com Claude e a API Claude
- [Status do sistema](https://docs.claude.com/pt/resources/status.md)
- [Получить API ключ](https://docs.claude.com/ru/api/admin-api/apikeys/get-api-key.md)
- [Список API ключей](https://docs.claude.com/ru/api/admin-api/apikeys/list-api-keys.md)
- [Обновить API ключи](https://docs.claude.com/ru/api/admin-api/apikeys/update-api-key.md)
- [Получить отчет об использовании Claude Code](https://docs.claude.com/ru/api/admin-api/claude-code/get-claude-code-usage-report.md): Retrieve daily aggregated usage metrics for Claude Code users.
Enables organizations to analyze developer productivity and build custom dashboards.
- [Создать приглашение](https://docs.claude.com/ru/api/admin-api/invites/create-invite.md)
- [Удалить приглашение](https://docs.claude.com/ru/api/admin-api/invites/delete-invite.md)
- [Получить приглашение](https://docs.claude.com/ru/api/admin-api/invites/get-invite.md)
- [Список приглашений](https://docs.claude.com/ru/api/admin-api/invites/list-invites.md)
- [Получить информацию об организации](https://docs.claude.com/ru/api/admin-api/organization/get-me.md)
- [Получить отчет о расходах](https://docs.claude.com/ru/api/admin-api/usage-cost/get-cost-report.md)
- [Получить отчет об использовании для API сообщений](https://docs.claude.com/ru/api/admin-api/usage-cost/get-messages-usage-report.md)
- [Получить пользователя](https://docs.claude.com/ru/api/admin-api/users/get-user.md)
- [Список пользователей](https://docs.claude.com/ru/api/admin-api/users/list-users.md)
- [Удалить пользователя](https://docs.claude.com/ru/api/admin-api/users/remove-user.md)
- [Обновить пользователя](https://docs.claude.com/ru/api/admin-api/users/update-user.md)
- [Добавить участника рабочего пространства](https://docs.claude.com/ru/api/admin-api/workspace_members/create-workspace-member.md)
- [Удалить участника рабочего пространства](https://docs.claude.com/ru/api/admin-api/workspace_members/delete-workspace-member.md)
- [Получить участника рабочего пространства](https://docs.claude.com/ru/api/admin-api/workspace_members/get-workspace-member.md)
- [Список участников рабочего пространства](https://docs.claude.com/ru/api/admin-api/workspace_members/list-workspace-members.md)
- [Обновить участника рабочего пространства](https://docs.claude.com/ru/api/admin-api/workspace_members/update-workspace-member.md)
- [Архивировать рабочее пространство](https://docs.claude.com/ru/api/admin-api/workspaces/archive-workspace.md)
- [Создать рабочее пространство](https://docs.claude.com/ru/api/admin-api/workspaces/create-workspace.md)
- [Получить рабочее пространство](https://docs.claude.com/ru/api/admin-api/workspaces/get-workspace.md)
- [Список рабочих пространств](https://docs.claude.com/ru/api/admin-api/workspaces/list-workspaces.md)
- [Обновить рабочее пространство](https://docs.claude.com/ru/api/admin-api/workspaces/update-workspace.md)
- [Обзор Admin API](https://docs.claude.com/ru/api/administration-api.md): Программно управляйте ресурсами вашей организации, включая участников организации, рабочие пространства и API ключи.
- [Beta заголовки](https://docs.claude.com/ru/api/beta-headers.md): Документация по использованию beta заголовков с Claude API
- [Отменить пакет сообщений](https://docs.claude.com/ru/api/canceling-message-batches.md): Batches may be canceled any time before processing ends. Once cancellation is initiated, the batch enters a `canceling` state, at which time the system may complete any in-progress, non-interruptible requests before finalizing cancellation.
The number of canceled requests is specified in `request_counts`. To determine which requests were canceled, check the individual results within the batch. Note that cancellation may not result in any canceled requests if they were non-interruptible.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [API аналитики Claude Code](https://docs.claude.com/ru/api/claude-code-analytics-api.md): Программный доступ к аналитике использования Claude Code вашей организации и метрикам продуктивности с помощью Admin API аналитики Claude Code.
- [Amazon Bedrock API](https://docs.claude.com/ru/api/claude-on-amazon-bedrock.md): Модели Claude от Anthropic теперь общедоступны через Amazon Bedrock.
- [Vertex AI API](https://docs.claude.com/ru/api/claude-on-vertex-ai.md): Модели Claude от Anthropic теперь общедоступны через [Vertex AI](https://cloud.google.com/vertex-ai).
- [Клиентские SDK](https://docs.claude.com/ru/api/client-sdks.md): Мы предоставляем клиентские библиотеки на нескольких популярных языках, которые упрощают работу с API Claude.
- [Создать пакет сообщений](https://docs.claude.com/ru/api/creating-message-batches.md): Send a batch of Message creation requests.
The Message Batches API can be used to process multiple Messages API requests at once. Once a Message Batch is created, it begins processing immediately. Batches can take up to 24 hours to complete.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Удалить пакет сообщений](https://docs.claude.com/ru/api/deleting-message-batches.md): Delete a Message Batch.
Message Batches can only be deleted once they've finished processing. If you'd like to delete an in-progress batch, you must first cancel it.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Ошибки](https://docs.claude.com/ru/api/errors.md)
- [Скачать файл](https://docs.claude.com/ru/api/files-content.md): Скачать содержимое файла, созданного Claude
- [Создать файл](https://docs.claude.com/ru/api/files-create.md): Загрузить файл
- [Удалить файл](https://docs.claude.com/ru/api/files-delete.md): Сделать файл недоступным через API
- [Список файлов](https://docs.claude.com/ru/api/files-list.md): Список файлов в рабочем пространстве
- [Получить метаданные файла](https://docs.claude.com/ru/api/files-metadata.md)
- [Получение помощи](https://docs.claude.com/ru/api/getting-help.md): Мы постарались предоставить ответы на наиболее часто задаваемые вопросы в этой документации. Однако, если вам нужна дополнительная техническая поддержка по использованию Claude, Claude API или любых наших продуктов, вы можете обратиться к нашей команде поддержки по адресу [support.claude.com](https://support.claude.com).
- [Обработка причин остановки](https://docs.claude.com/ru/api/handling-stop-reasons.md): Узнайте, как правильно обрабатывать различные значения stop_reason в ответах API Messages от Claude для создания надежных приложений.
- [IP-адреса](https://docs.claude.com/ru/api/ip-addresses.md): Сервисы Anthropic используют фиксированные IP-адреса как для входящих, так и для исходящих соединений. Вы можете использовать эти адреса для настройки правил брандмауэра для безопасного доступа к Claude API и Консоли. Эти адреса не будут изменены без уведомления.
- [Список пакетов сообщений](https://docs.claude.com/ru/api/listing-message-batches.md): List all Message Batches within a Workspace. Most recently created batches are returned first.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Сообщения](https://docs.claude.com/ru/api/messages.md): Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless multi-turn conversations.
Learn more about the Messages API in our [user guide](/en/docs/initial-setup)
- [Примеры пакетов сообщений](https://docs.claude.com/ru/api/messages-batch-examples.md): Примеры использования API пакетов сообщений
- [Подсчет токенов сообщения](https://docs.claude.com/ru/api/messages-count-tokens.md): Count the number of tokens in a Message.
The Token Count API can be used to count the number of tokens in a Message, including tools, images, and documents, without creating it.
Learn more about token counting in our [user guide](/en/docs/build-with-claude/token-counting)
- [Примеры сообщений](https://docs.claude.com/ru/api/messages-examples.md): Примеры запросов и ответов для Messages API
- [Миграция с Text Completions](https://docs.claude.com/ru/api/migrating-from-text-completions-to-messages.md): Миграция с Text Completions на Messages
- [Получить модель](https://docs.claude.com/ru/api/models.md): Get a specific model.
The Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.
- [Список моделей](https://docs.claude.com/ru/api/models-list.md): List available models.
The Models API response can be used to determine which models are available for use in the API. More recently released models are listed first.
- [Совместимость с OpenAI SDK](https://docs.claude.com/ru/api/openai-sdk.md): Anthropic предоставляет слой совместимости, который позволяет использовать OpenAI SDK для тестирования Claude API. С помощью нескольких изменений в коде вы можете быстро оценить возможности моделей Anthropic.
- [Обзор](https://docs.claude.com/ru/api/overview.md): Обзор API Claude
- [Генерация промпта](https://docs.claude.com/ru/api/prompt-tools-generate.md): Генерация хорошо написанного промпта
- [Улучшить промпт](https://docs.claude.com/ru/api/prompt-tools-improve.md): Создать новый и улучшенный промпт на основе обратной связи
- [Создание шаблона промпта](https://docs.claude.com/ru/api/prompt-tools-templatize.md): Создание шаблона промпта путем идентификации и извлечения переменных
- [Ограничения скорости](https://docs.claude.com/ru/api/rate-limits.md): Для предотвращения злоупотреблений и управления пропускной способностью нашего API мы внедрили ограничения на то, сколько организация может использовать Claude API.
- [Получить результаты пакета сообщений](https://docs.claude.com/ru/api/retrieving-message-batch-results.md): Streams the results of a Message Batch as a `.jsonl` file.
Each line in the file is a JSON object containing the result of a single request in the Message Batch. Results are not guaranteed to be in the same order as requests. Use the `custom_id` field to match results to requests.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Получить пакет сообщений](https://docs.claude.com/ru/api/retrieving-message-batches.md): This endpoint is idempotent and can be used to poll for Message Batch completion. To access the results of a Message Batch, make a request to the `results_url` field in the response.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Уровни обслуживания](https://docs.claude.com/ru/api/service-tiers.md): Различные уровни обслуживания позволяют вам балансировать доступность, производительность и предсказуемые затраты в зависимости от потребностей вашего приложения.
- [Поддерживаемые регионы](https://docs.claude.com/ru/api/supported-regions.md): Вот страны, регионы и территории, из которых мы в настоящее время можем поддерживать доступ:
- [API использования и стоимости](https://docs.claude.com/ru/api/usage-cost-api.md): Программно получайте доступ к данным об использовании API и стоимости вашей организации с помощью Admin API для использования и стоимости.
- [Версии](https://docs.claude.com/ru/api/versioning.md): При выполнении API-запросов вы должны отправлять заголовок запроса `anthropic-version`. Например, `anthropic-version: 2023-06-01`. Если вы используете наши [клиентские SDK](/ru/api/client-sdks), это обрабатывается автоматически.
- [Глоссарий](https://docs.claude.com/ru/docs/about-claude/glossary.md): Эти концепции не являются уникальными для языковых моделей Anthropic, но мы представляем краткое изложение ключевых терминов ниже.
- [Устаревание моделей](https://docs.claude.com/ru/docs/about-claude/model-deprecations.md)
- [Выбор подходящей модели](https://docs.claude.com/ru/docs/about-claude/models/choosing-a-model.md): Выбор оптимальной модели Claude для вашего приложения включает балансирование трех ключевых факторов: возможности, скорость и стоимость. Это руководство поможет вам принять обоснованное решение на основе ваших конкретных требований.
- [Миграция на Claude 4](https://docs.claude.com/ru/docs/about-claude/models/migrating-to-claude-4.md): Руководство по миграции с моделей Claude 3.7 на модели Claude 4
- [Обзор моделей](https://docs.claude.com/ru/docs/about-claude/models/overview.md): Claude — это семейство современных больших языковых моделей, разработанных Anthropic. Это руководство знакомит с нашими моделями и сравнивает их производительность с устаревшими моделями.
- [Цены](https://docs.claude.com/ru/docs/about-claude/pricing.md): Узнайте о структуре ценообразования Anthropic для моделей и функций
- [Модерация контента](https://docs.claude.com/ru/docs/about-claude/use-case-guides/content-moderation.md): Модерация контента является критически важным аспектом поддержания безопасной, уважительной и продуктивной среды в цифровых приложениях. В этом руководстве мы обсудим, как Claude может использоваться для модерации контента в вашем цифровом приложении.
- [Агент службы поддержки клиентов](https://docs.claude.com/ru/docs/about-claude/use-case-guides/customer-support-chat.md): Это руководство объясняет, как использовать продвинутые разговорные возможности Claude для обработки запросов клиентов в режиме реального времени, обеспечивая круглосуточную поддержку, сокращая время ожидания и управляя большими объемами запросов поддержки с точными ответами и позитивными взаимодействиями.
- [Правовое резюмирование](https://docs.claude.com/ru/docs/about-claude/use-case-guides/legal-summarization.md): Это руководство показывает, как использовать передовые возможности обработки естественного языка Claude для эффективного резюмирования правовых документов, извлечения ключевой информации и ускорения правовых исследований. С Claude вы можете оптимизировать рассмотрение контрактов, подготовку к судебным разбирательствам и регулятивную работу, экономя время и обеспечивая точность в ваших правовых процессах.
- [Руководства по распространенным сценариям использования](https://docs.claude.com/ru/docs/about-claude/use-case-guides/overview.md)
- [Маршрутизация заявок](https://docs.claude.com/ru/docs/about-claude/use-case-guides/ticket-routing.md): Это руководство объясняет, как использовать передовые возможности понимания естественного языка Claude для классификации заявок службы поддержки клиентов в масштабе на основе намерений клиентов, срочности, приоритизации, профиля клиента и многого другого.
- [Дополнение для Google Sheets](https://docs.claude.com/ru/docs/agents-and-tools/claude-for-sheets.md): Расширение [Claude for Sheets](https://workspace.google.com/marketplace/app/claude%5Ffor%5Fsheets/909417792257) интегрирует Claude в Google Sheets, позволяя выполнять взаимодействия с Claude непосредственно в ячейках.
- [MCP коннектор](https://docs.claude.com/ru/docs/agents-and-tools/mcp-connector.md)
- [Удаленные MCP серверы](https://docs.claude.com/ru/docs/agents-and-tools/remote-mcp-servers.md): Удаленные MCP серверы
- [Инструмент Bash](https://docs.claude.com/ru/docs/agents-and-tools/tool-use/bash-tool.md): Инструмент bash позволяет Claude выполнять команды оболочки в постоянной сессии bash, обеспечивая системные операции, выполнение скриптов и автоматизацию командной строки.
- [Инструмент выполнения кода](https://docs.claude.com/ru/docs/agents-and-tools/tool-use/code-execution-tool.md)
- [Инструмент использования компьютера](https://docs.claude.com/ru/docs/agents-and-tools/tool-use/computer-use-tool.md): Claude может взаимодействовать с компьютерными средами через инструмент использования компьютера, который предоставляет возможности создания скриншотов и управления мышью/клавиатурой для автономного взаимодействия с рабочим столом.
- [Детализированная потоковая передача инструментов](https://docs.claude.com/ru/docs/agents-and-tools/tool-use/fine-grained-tool-streaming.md): Детализированная потоковая передача инструментов
- [Как реализовать использование инструментов](https://docs.claude.com/ru/docs/agents-and-tools/tool-use/implement-tool-use.md): Руководство по реализации использования инструментов с Claude, включая выбор модели, определение инструментов, управление выводом и обработку ошибок.
- [Использование инструментов с Claude](https://docs.claude.com/ru/docs/agents-and-tools/tool-use/overview.md)
- [Инструмент текстового редактора](https://docs.claude.com/ru/docs/agents-and-tools/tool-use/text-editor-tool.md): Позволяет Claude просматривать и изменять текстовые файлы, помогая отлаживать, исправлять и улучшать код или другие текстовые документы.
- [Токен-эффективное использование инструментов](https://docs.claude.com/ru/docs/agents-and-tools/tool-use/token-efficient-tool-use.md)
- [Инструмент веб-загрузки](https://docs.claude.com/ru/docs/agents-and-tools/tool-use/web-fetch-tool.md): Инструмент веб-загрузки позволяет Claude получать полное содержимое с указанных веб-страниц и PDF-документов.
- [Инструмент веб-поиска](https://docs.claude.com/ru/docs/agents-and-tools/tool-use/web-search-tool.md): Инструмент веб-поиска дает Claude прямой доступ к веб-контенту в реальном времени, позволяя отвечать на вопросы с актуальной информацией.
- [Пакетная обработка](https://docs.claude.com/ru/docs/build-with-claude/batch-processing.md)
- [Цитирование](https://docs.claude.com/ru/docs/build-with-claude/citations.md): Цитирование
- [Контекстные окна](https://docs.claude.com/ru/docs/build-with-claude/context-windows.md)
- [Эмбеддинги](https://docs.claude.com/ru/docs/build-with-claude/embeddings.md): Текстовые эмбеддинги — это числовые представления текста, которые позволяют измерять семантическое сходство. Это руководство знакомит с эмбеддингами, их применением и тем, как использовать модели эмбеддингов для таких задач, как поиск, рекомендации и обнаружение аномалий.
- [Создание с расширенным мышлением](https://docs.claude.com/ru/docs/build-with-claude/extended-thinking.md)
- [Files API](https://docs.claude.com/ru/docs/build-with-claude/files.md)
- [Многоязычная поддержка](https://docs.claude.com/ru/docs/build-with-claude/multilingual-support.md): Claude превосходно справляется с задачами на разных языках, сохраняя высокую производительность на разных языках относительно английского.
- [Обзор функций](https://docs.claude.com/ru/docs/build-with-claude/overview.md): Изучите продвинутые функции и возможности Claude.
- [Поддержка PDF](https://docs.claude.com/ru/docs/build-with-claude/pdf-support.md): Обрабатывайте PDF с помощью Claude. Извлекайте текст, анализируйте диаграммы и понимайте визуальное содержимое ваших документов.
- [Кеширование промптов](https://docs.claude.com/ru/docs/build-with-claude/prompt-caching.md)
- [Будьте ясными, прямыми и подробными](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/be-clear-and-direct.md)
- [Позвольте Claude думать (цепочка рассуждений) для повышения эффективности](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/chain-of-thought.md)
- [Цепочка сложных промптов для повышения эффективности](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/chain-prompts.md)
- [Лучшие практики промпт-инжиниринга для Claude 4](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/claude-4-best-practices.md): Руководство по специфическим техникам промпт-инжиниринга для моделей Claude 4 для достижения оптимальных результатов в ваших приложениях.
- [Советы по расширенному мышлению](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/extended-thinking-tips.md)
- [Советы по работе с длинным контекстом](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/long-context-tips.md)
- [Используйте примеры (мультишотовый промптинг) для управления поведением Claude](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/multishot-prompting.md)
- [Обзор инженерии промптов](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/overview.md): Обзор инженерии промптов
- [Предзаполнение ответа Claude для большего контроля над выводом](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/prefill-claudes-response.md)
- [Автоматическая генерация шаблонов промптов первого черновика](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/prompt-generator.md)
- [Используйте наш улучшитель промптов для оптимизации ваших промптов](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/prompt-improver.md): Используйте наш улучшитель промптов для оптимизации ваших промптов
- [Использование шаблонов промптов и переменных](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/prompt-templates-and-variables.md)
- [Назначение роли Claude с помощью системного промпта](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/system-prompts.md)
- [Используйте XML-теги для структурирования ваших промптов](https://docs.claude.com/ru/docs/build-with-claude/prompt-engineering/use-xml-tags.md)
- [Результаты поиска](https://docs.claude.com/ru/docs/build-with-claude/search-results.md): Включите естественные цитирования для RAG-приложений, предоставляя результаты поиска с атрибуцией источников
- [Потоковые сообщения](https://docs.claude.com/ru/docs/build-with-claude/streaming.md): Узнайте, как использовать потоковую передачу для получения инкрементальных ответов от Claude с помощью server-sent events (SSE).
- [Подсчет токенов](https://docs.claude.com/ru/docs/build-with-claude/token-counting.md): Определите количество токенов в сообщении перед отправкой Claude, чтобы принимать обоснованные решения о ваших промптах и использовании.
- [Зрение](https://docs.claude.com/ru/docs/build-with-claude/vision.md): Семейства моделей Claude 3 и 4 поставляются с новыми возможностями зрения, которые позволяют Claude понимать и анализировать изображения, открывая захватывающие возможности для мультимодального взаимодействия.
- [Claude Code на Amazon Bedrock](https://docs.claude.com/ru/docs/claude-code/amazon-bedrock.md): Узнайте о настройке Claude Code через Amazon Bedrock, включая установку, конфигурацию IAM и устранение неполадок.
- [Аналитика](https://docs.claude.com/ru/docs/claude-code/analytics.md): Просматривайте подробную информацию об использовании и метрики производительности для развертывания Claude Code в вашей организации.
- [Справочник CLI](https://docs.claude.com/ru/docs/claude-code/cli-reference.md): Полный справочник по интерфейсу командной строки Claude Code, включая команды и флаги.
- [Общие рабочие процессы](https://docs.claude.com/ru/docs/claude-code/common-workflows.md): Изучите общие рабочие процессы с Claude Code.
- [Эффективное управление расходами](https://docs.claude.com/ru/docs/claude-code/costs.md): Узнайте, как отслеживать и оптимизировать использование токенов и расходы при использовании Claude Code.
- [Использование данных](https://docs.claude.com/ru/docs/claude-code/data-usage.md): Узнайте о политике использования данных Anthropic для Claude
- [Контейнеры разработки](https://docs.claude.com/ru/docs/claude-code/devcontainer.md): Узнайте о контейнере разработки Claude Code для команд, которым нужны согласованные, безопасные среды.
- [Claude Code GitHub Actions](https://docs.claude.com/ru/docs/claude-code/github-actions.md): Узнайте об интеграции Claude Code в ваш рабочий процесс разработки с помощью Claude Code GitHub Actions
- [Claude Code GitLab CI/CD](https://docs.claude.com/ru/docs/claude-code/gitlab-ci-cd.md): Узнайте об интеграции Claude Code в ваш рабочий процесс разработки с GitLab CI/CD
- [Claude Code на Google Vertex AI](https://docs.claude.com/ru/docs/claude-code/google-vertex-ai.md): Узнайте о настройке Claude Code через Google Vertex AI, включая установку, конфигурацию IAM и устранение неполадок.
- [Справочник по хукам](https://docs.claude.com/ru/docs/claude-code/hooks.md): Эта страница предоставляет справочную документацию по реализации хуков в Claude Code.
- [Начало работы с хуками Claude Code](https://docs.claude.com/ru/docs/claude-code/hooks-guide.md): Узнайте, как настраивать и расширять поведение Claude Code, регистрируя команды оболочки
- [Управление идентификацией и доступом](https://docs.claude.com/ru/docs/claude-code/iam.md): Узнайте, как настроить аутентификацию пользователей, авторизацию и контроль доступа для Claude Code в вашей организации.
- [Добавьте Claude Code в вашу IDE](https://docs.claude.com/ru/docs/claude-code/ide-integrations.md): Узнайте, как добавить Claude Code в вашу любимую IDE
- [Интерактивный режим](https://docs.claude.com/ru/docs/claude-code/interactive-mode.md): Полный справочник по сочетаниям клавиш, режимам ввода и интерактивным функциям в сессиях Claude Code.
- [Правовые вопросы и соответствие требованиям](https://docs.claude.com/ru/docs/claude-code/legal-and-compliance.md): Правовые соглашения, сертификаты соответствия и информация о безопасности для Claude Code.
- [Конфигурация LLM шлюза](https://docs.claude.com/ru/docs/claude-code/llm-gateway.md): Узнайте, как настроить Claude Code с решениями LLM шлюза, включая настройку LiteLLM, методы аутентификации и корпоративные функции, такие как отслеживание использования и управление бюджетом.
- [Подключение Claude Code к инструментам через MCP](https://docs.claude.com/ru/docs/claude-code/mcp.md): Узнайте, как подключить Claude Code к вашим инструментам с помощью Model Context Protocol.
- [Управление памятью Claude](https://docs.claude.com/ru/docs/claude-code/memory.md): Узнайте, как управлять памятью Claude Code между сессиями с различными местоположениями памяти и лучшими практиками.
- [Конфигурация модели](https://docs.claude.com/ru/docs/claude-code/model-config.md): Изучите конфигурацию модели Claude Code, включая псевдонимы моделей, такие как `opusplan`
- [Мониторинг](https://docs.claude.com/ru/docs/claude-code/monitoring-usage.md): Узнайте, как включить и настроить OpenTelemetry для Claude Code.
- [Корпоративная конфигурация сети](https://docs.claude.com/ru/docs/claude-code/network-config.md): Настройте Claude Code для корпоративных сред с прокси-серверами, пользовательскими центрами сертификации (CA) и взаимной аутентификацией Transport Layer Security (mTLS).
- [Стили вывода](https://docs.claude.com/ru/docs/claude-code/output-styles.md): Адаптируйте Claude Code для использования за пределами разработки программного обеспечения
- [Обзор Claude Code](https://docs.claude.com/ru/docs/claude-code/overview.md): Узнайте о Claude Code, агентном инструменте для кодирования от Anthropic, который работает в вашем терминале и помогает превращать идеи в код быстрее, чем когда-либо прежде.
- [Быстрый старт](https://docs.claude.com/ru/docs/claude-code/quickstart.md): Добро пожаловать в Claude Code!
- [Пользовательские инструменты](https://docs.claude.com/ru/docs/claude-code/sdk/custom-tools.md): Создавайте и интегрируйте пользовательские инструменты для расширения функциональности Claude Code SDK
- [Изменение системных промптов](https://docs.claude.com/ru/docs/claude-code/sdk/modifying-system-prompts.md): Узнайте, как настроить поведение Claude, изменяя системные промпты с помощью трех подходов - стили вывода, appendSystemPrompt и customSystemPrompt.
- [Отслеживание затрат и использования](https://docs.claude.com/ru/docs/claude-code/sdk/sdk-cost-tracking.md): Понимание и отслеживание использования токенов для выставления счетов в Claude Code SDK
- [Режим без интерфейса](https://docs.claude.com/ru/docs/claude-code/sdk/sdk-headless.md): Запуск Claude Code программно без интерактивного пользовательского интерфейса
- [MCP в SDK](https://docs.claude.com/ru/docs/claude-code/sdk/sdk-mcp.md): Расширьте Claude Code с помощью пользовательских инструментов, используя серверы Model Context Protocol
- [Обзор](https://docs.claude.com/ru/docs/claude-code/sdk/sdk-overview.md): Создавайте пользовательских ИИ-агентов с помощью Claude Code SDK
- [Управление разрешениями](https://docs.claude.com/ru/docs/claude-code/sdk/sdk-permissions.md): Контроль использования инструментов и разрешений в Claude Code SDK
- [Справочник по Python SDK](https://docs.claude.com/ru/docs/claude-code/sdk/sdk-python.md): Полный справочник по API для Claude Code Python SDK, включая все функции, типы и классы.
- [Управление сессиями](https://docs.claude.com/ru/docs/claude-code/sdk/sdk-sessions.md): Понимание того, как Claude Code SDK обрабатывает сессии и возобновление сессий
- [Слэш-команды в SDK](https://docs.claude.com/ru/docs/claude-code/sdk/sdk-slash-commands.md): Узнайте, как использовать слэш-команды для управления сессиями Claude Code через SDK
- [Справочник TypeScript SDK](https://docs.claude.com/ru/docs/claude-code/sdk/sdk-typescript.md): Полный справочник API для TypeScript SDK Claude Code, включая все функции, типы и интерфейсы.
- [Потоковый ввод](https://docs.claude.com/ru/docs/claude-code/sdk/streaming-vs-single-mode.md): Понимание двух режимов ввода для Claude Code SDK и когда использовать каждый
- [Субагенты в SDK](https://docs.claude.com/ru/docs/claude-code/sdk/subagents.md): Работа с субагентами в Claude Code SDK
- [Списки дел](https://docs.claude.com/ru/docs/claude-code/sdk/todo-tracking.md): Отслеживайте и отображайте задачи с помощью Claude Code SDK для организованного управления задачами
- [Безопасность](https://docs.claude.com/ru/docs/claude-code/security.md): Узнайте о мерах безопасности Claude Code и лучших практиках для безопасного использования.
- [Настройки Claude Code](https://docs.claude.com/ru/docs/claude-code/settings.md): Настройте Claude Code с помощью глобальных настроек уровня проекта и переменных окружения.
- [Настройка Claude Code](https://docs.claude.com/ru/docs/claude-code/setup.md): Установите, аутентифицируйтесь и начните использовать Claude Code на вашей машине разработки.
- [Команды слэш](https://docs.claude.com/ru/docs/claude-code/slash-commands.md): Управляйте поведением Claude во время интерактивной сессии с помощью команд слэш.
- [Конфигурация строки состояния](https://docs.claude.com/ru/docs/claude-code/statusline.md): Создайте пользовательскую строку состояния для Claude Code для отображения контекстной информации
- [Субагенты](https://docs.claude.com/ru/docs/claude-code/sub-agents.md): Создавайте и используйте специализированных ИИ-субагентов в Claude Code для задачно-ориентированных рабочих процессов и улучшенного управления контекстом.
- [Оптимизируйте настройку вашего терминала](https://docs.claude.com/ru/docs/claude-code/terminal-config.md): Claude Code работает лучше всего, когда ваш терминал правильно настроен. Следуйте этим рекомендациям для оптимизации вашего опыта.
- [Обзор корпоративного развертывания](https://docs.claude.com/ru/docs/claude-code/third-party-integrations.md): Узнайте, как Claude Code может интегрироваться с различными сторонними сервисами и инфраструктурой для удовлетворения требований корпоративного развертывания.
- [Устранение неполадок](https://docs.claude.com/ru/docs/claude-code/troubleshooting.md): Откройте для себя решения распространенных проблем с установкой и использованием Claude Code.
- [Начало работы с Claude](https://docs.claude.com/ru/docs/get-started.md): Сделайте свой первый API-вызов к Claude и создайте простого помощника для веб-поиска
- [Введение в Claude](https://docs.claude.com/ru/docs/intro.md): Claude — это высокопроизводительная, надежная и интеллектуальная AI-платформа, созданная Anthropic. Claude превосходно справляется с задачами, связанными с языком, рассуждениями, анализом, программированием и многим другим.
- [Политика конфиденциальности Anthropic](https://docs.claude.com/ru/docs/legal-center/privacy.md)
- [Безопасность и соответствие требованиям](https://docs.claude.com/ru/docs/legal-center/security-compliance.md)
- [Создание с Claude](https://docs.claude.com/ru/docs/overview.md): Claude — это семейство [высокопроизводительных и интеллектуальных ИИ-моделей](/ru/docs/about-claude/models), созданных Anthropic. Хотя Claude мощный и расширяемый, он также является самым надежным и безопасным доступным ИИ. Он следует критически важным протоколам, делает меньше ошибок и устойчив к взломам — позволяя [корпоративным клиентам](https://claude.com/customers) создавать самые безопасные ИИ-приложения в масштабе.
- [Определите критерии успеха](https://docs.claude.com/ru/docs/test-and-evaluate/define-success.md)
- [Создание надежных эмпирических оценок](https://docs.claude.com/ru/docs/test-and-evaluate/develop-tests.md): Узнайте, как разрабатывать тестовые случаи для измерения производительности LLM в соответствии с вашими критериями успеха.
- [Использование инструмента оценки](https://docs.claude.com/ru/docs/test-and-evaluate/eval-tool.md): В [Claude Console](https://console.anthropic.com/dashboard) есть **инструмент оценки**, который позволяет тестировать ваши промпты в различных сценариях.
- [null](https://docs.claude.com/ru/docs/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals.md)
- [Повышение согласованности вывода (режим JSON)](https://docs.claude.com/ru/docs/test-and-evaluate/strengthen-guardrails/increase-consistency.md)
- [Поддержание характера Claude с помощью ролевых промптов и предварительного заполнения](https://docs.claude.com/ru/docs/test-and-evaluate/strengthen-guardrails/keep-claude-in-character.md)
- [Предотвращение взлома и инъекций промптов](https://docs.claude.com/ru/docs/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks.md)
- [Уменьшение галлюцинаций](https://docs.claude.com/ru/docs/test-and-evaluate/strengthen-guardrails/reduce-hallucinations.md)
- [Снижение задержки](https://docs.claude.com/ru/docs/test-and-evaluate/strengthen-guardrails/reduce-latency.md)
- [Уменьшение утечки промпта](https://docs.claude.com/ru/docs/test-and-evaluate/strengthen-guardrails/reduce-prompt-leak.md)
- [null](https://docs.claude.com/ru/home.md)
- [Платформа разработчика Claude](https://docs.claude.com/ru/release-notes/api.md): Следите за обновлениями API Claude и консоли Claude.
- [Приложения Claude](https://docs.claude.com/ru/release-notes/claude-apps.md): Следите за обновлениями приложений Claude от Anthropic.
- [Claude Code](https://docs.claude.com/ru/release-notes/claude-code.md): Будьте в курсе последних функций и улучшений Claude Code
- [Обзор](https://docs.claude.com/ru/release-notes/overview.md): Следите за обновлениями продуктов и сервисов Anthropic.
- [Системные промпты](https://docs.claude.com/ru/release-notes/system-prompts.md): Смотрите обновления основных системных промптов на [Claude.ai](https://www.claude.ai) и в приложениях Claude для [iOS](http://anthropic.com/ios) и [Android](http://anthropic.com/android).
- [Системная карта Claude 3.7](https://docs.claude.com/ru/resources/claude-3-7-system-card.md)
- [Карточка модели Claude 3](https://docs.claude.com/ru/resources/claude-3-model-card.md)
- [Системная карта Claude 4](https://docs.claude.com/ru/resources/claude-4-system-card.md): Системная карта Claude 4
- [Курсы Anthropic](https://docs.claude.com/ru/resources/courses.md)
- [null](https://docs.claude.com/ru/resources/overview.md)
- [Адаптивный редактор](https://docs.claude.com/ru/resources/prompt-library/adaptive-editor.md): Переписывает текст согласно инструкциям пользователя, например, с другим тоном, аудиторией или стилем.
- [Анализатор кодов аэропортов](https://docs.claude.com/ru/resources/prompt-library/airport-code-analyst.md): Найти и извлечь коды аэропортов из текста.
- [Инопланетный антрополог](https://docs.claude.com/ru/resources/prompt-library/alien-anthropologist.md): Анализируйте человеческую культуру и обычаи с точки зрения инопланетного антрополога.
- [Алхимик аллитерации](https://docs.claude.com/ru/resources/prompt-library/alliteration-alchemist.md): Создавайте аллитеративные фразы и предложения для любой заданной темы.
- [Трансляции Babel](https://docs.claude.com/ru/resources/prompt-library/babels-broadcasts.md): Создавайте убедительные твиты с объявлениями о продуктах на 10 самых распространенных языках мира.
- [Создатель бренда](https://docs.claude.com/ru/resources/prompt-library/brand-builder.md): Создайте дизайн-бриф для целостной идентичности бренда.
- [Карьерный коуч](https://docs.claude.com/ru/resources/prompt-library/career-coach.md): Участвуйте в ролевых разговорах с AI карьерным коучем.
- [Ссылайтесь на источники](https://docs.claude.com/ru/resources/prompt-library/cite-your-sources.md): Получайте ответы на вопросы о содержании документа с соответствующими цитатами, подтверждающими ответ.
- [Разъяснитель кода](https://docs.claude.com/ru/resources/prompt-library/code-clarifier.md): Упрощайте и объясняйте сложный код простым языком.
- [Консультант по коду](https://docs.claude.com/ru/resources/prompt-library/code-consultant.md): Предложите улучшения для оптимизации производительности Python кода.
- [Корпоративный ясновидящий](https://docs.claude.com/ru/resources/prompt-library/corporate-clairvoyant.md): Извлекайте инсайты, выявляйте риски и выделяйте ключевую информацию из длинных корпоративных отчетов в одну записку.
- [Космические Нажатия Клавиш](https://docs.claude.com/ru/resources/prompt-library/cosmic-keystrokes.md): Создайте интерактивную игру на скорость печати в одном HTML файле с боковой прокруткой и стилизацией Tailwind CSS.
- [Конвертер CSV](https://docs.claude.com/ru/resources/prompt-library/csv-converter.md): Преобразование данных из различных форматов (JSON, XML и т.д.) в правильно отформатированные CSV файлы.
- [Кулинарный творец](https://docs.claude.com/ru/resources/prompt-library/culinary-creator.md): Предлагайте идеи рецептов на основе доступных ингредиентов пользователя и диетических предпочтений.
- [Организатор данных](https://docs.claude.com/ru/resources/prompt-library/data-organizer.md): Превращайте неструктурированный текст в индивидуальные JSON-таблицы.
- [Декодер инструкций](https://docs.claude.com/ru/resources/prompt-library/direction-decoder.md): Преобразуйте естественный язык в пошаговые инструкции.
- [Толкователь снов](https://docs.claude.com/ru/resources/prompt-library/dream-interpreter.md): Предлагайте интерпретации и понимание символизма снов пользователя.
- [Оценщик эффективности](https://docs.claude.com/ru/resources/prompt-library/efficiency-estimator.md): Вычисляет временную сложность функций и алгоритмов.
- [Извлекатель электронной почты](https://docs.claude.com/ru/resources/prompt-library/email-extractor.md): Извлекайте адреса электронной почты из документа в список в формате JSON.
- [Кодировщик эмодзи](https://docs.claude.com/ru/resources/prompt-library/emoji-encoder.md): Преобразуйте обычный текст в веселые и выразительные сообщения с эмодзи.
- [Навигатор этических дилемм](https://docs.claude.com/ru/resources/prompt-library/ethical-dilemma-navigator.md): Помогите пользователю обдумать сложные этические дилеммы и предоставьте различные точки зрения.
- [Эксперт по формулам Excel](https://docs.claude.com/ru/resources/prompt-library/excel-formula-expert.md): Создавайте формулы Excel на основе описанных пользователем вычислений или манипуляций с данными.
- [Фабрикатор функций](https://docs.claude.com/ru/resources/prompt-library/function-fabricator.md): Создание Python функций на основе детальных спецификаций.
- [Футуристический модный консультант](https://docs.claude.com/ru/resources/prompt-library/futuristic-fashion-advisor.md): Предлагайте авангардные модные тренды и стили для конкретных предпочтений пользователя.
- [Git gud](https://docs.claude.com/ru/resources/prompt-library/git-gud.md): Генерируйте подходящие команды Git на основе описанных пользователем действий контроля версий.
- [Скриптер Google apps](https://docs.claude.com/ru/resources/prompt-library/google-apps-scripter.md): Генерирует скрипты Google Apps для выполнения задач на основе требований пользователя.
- [Гуру оценивания](https://docs.claude.com/ru/resources/prompt-library/grading-guru.md): Сравнивайте и оценивайте качество письменных текстов на основе определенных пользователем критериев и стандартов.
- [Джинн грамматики](https://docs.claude.com/ru/resources/prompt-library/grammar-genie.md): Преобразуйте грамматически неправильные предложения в правильный английский язык.
- [Hal - юмористический помощник](https://docs.claude.com/ru/resources/prompt-library/hal-the-humorous-helper.md): Общайтесь с знающим ИИ, у которого есть саркастическая сторона.
- [Объяснитель идиом](https://docs.claude.com/ru/resources/prompt-library/idiom-illuminator.md): Объясните значение и происхождение распространенных идиом и пословиц.
- [Создатель вопросов для интервью](https://docs.claude.com/ru/resources/prompt-library/interview-question-crafter.md): Генерируйте вопросы для интервью.
- [Легенда LaTeX](https://docs.claude.com/ru/resources/prompt-library/latex-legend.md): Пишите документы LaTeX, генерируя код для математических уравнений, таблиц и многого другого.
- [Планировщик уроков](https://docs.claude.com/ru/resources/prompt-library/lesson-planner.md): Создавайте подробные планы уроков по любому предмету.
- [Библиотека промптов](https://docs.claude.com/ru/resources/prompt-library/library.md)
- [Главный модератор](https://docs.claude.com/ru/resources/prompt-library/master-moderator.md): Оценивайте пользовательские вводы на предмет потенциально вредного или незаконного контента.
- [Секретарь собрания](https://docs.claude.com/ru/resources/prompt-library/meeting-scribe.md): Преобразуйте собрания в краткие резюме, включающие темы обсуждения, ключевые выводы и пункты действий.
- [Мастер служебных записок](https://docs.claude.com/ru/resources/prompt-library/memo-maestro.md): Составляйте исчерпывающие служебные записки компании на основе ключевых пунктов.
- [Наставник по осознанности](https://docs.claude.com/ru/resources/prompt-library/mindfulness-mentor.md): Проведите пользователя через упражнения и техники осознанности для снижения стресса.
- [Колоризатор настроения](https://docs.claude.com/ru/resources/prompt-library/mood-colorizer.md): Преобразование текстовых описаний настроений в соответствующие HEX-коды.
- [Мотивационная муза](https://docs.claude.com/ru/resources/prompt-library/motivational-muse.md): Предоставляйте персонализированные мотивационные сообщения и аффирмации на основе пользовательского ввода.
- [Создатель неологизмов](https://docs.claude.com/ru/resources/prompt-library/neologism-creator.md): Изобретайте новые слова и предоставляйте их определения на основе концепций или идей, предоставленных пользователем.
- [Размышляющий о перспективах](https://docs.claude.com/ru/resources/prompt-library/perspectives-ponderer.md): Взвесьте плюсы и минусы темы, предоставленной пользователем.
- [Философские размышления](https://docs.claude.com/ru/resources/prompt-library/philosophical-musings.md): Участвуйте в глубоких философских дискуссиях и мысленных экспериментах.
- [Очиститель персональных данных](https://docs.claude.com/ru/resources/prompt-library/pii-purifier.md): Автоматически обнаруживает и удаляет персонально идентифицируемую информацию (PII) из текстовых документов.
- [Полиглот суперсилы](https://docs.claude.com/ru/resources/prompt-library/polyglot-superpowers.md): Переводите текст с любого языка на любой язык.
- [Поэт портманто](https://docs.claude.com/ru/resources/prompt-library/portmanteau-poet.md): Смешайте два слова вместе, чтобы создать новое, осмысленное портманто.
- [Профессиональное именование продуктов](https://docs.claude.com/ru/resources/prompt-library/product-naming-pro.md): Создавайте запоминающиеся названия продуктов на основе описаний и ключевых слов.
- [Полировщик прозы](https://docs.claude.com/ru/resources/prompt-library/prose-polisher.md): Улучшайте и совершенствуйте письменный контент с помощью продвинутых техник редактирования и предложений.
- [Каламбурщик](https://docs.claude.com/ru/resources/prompt-library/pun-dit.md): Генерируйте умные каламбуры и игру слов на основе любой заданной темы.
- [Исправитель ошибок Python](https://docs.claude.com/ru/resources/prompt-library/python-bug-buster.md): Обнаружение и исправление ошибок в коде Python.
- [Классификатор отзывов](https://docs.claude.com/ru/resources/prompt-library/review-classifier.md): Категоризируйте отзывы в предварительно заданные теги и категории.
- [Загадай мне это](https://docs.claude.com/ru/resources/prompt-library/riddle-me-this.md): Создавайте загадки и направляйте пользователя к решениям.
- [Симулятор научно-фантастических сценариев](https://docs.claude.com/ru/resources/prompt-library/sci-fi-scenario-simulator.md): Обсуждайте с пользователем различные научно-фантастические сценарии и связанные с ними вызовы и соображения.
- [Упроститель для второго класса](https://docs.claude.com/ru/resources/prompt-library/second-grade-simplifier.md): Делайте сложный текст понятным для юных учеников.
- [Знаток сравнений](https://docs.claude.com/ru/resources/prompt-library/simile-savant.md): Генерируйте сравнения из простых описаний.
- [Сократический мудрец](https://docs.claude.com/ru/resources/prompt-library/socratic-sage.md): Ведите беседу в сократическом стиле на заданную пользователем тему.
- [Волшебник электронных таблиц](https://docs.claude.com/ru/resources/prompt-library/spreadsheet-sorcerer.md): Генерируйте CSV электронные таблицы с различными типами данных.
- [SQL волшебник](https://docs.claude.com/ru/resources/prompt-library/sql-sorcerer.md): Преобразуйте обычный язык в SQL запросы.
- [Помощник в создании историй](https://docs.claude.com/ru/resources/prompt-library/storytelling-sidekick.md): Совместно создавайте увлекательные истории с пользователем, предлагая неожиданные повороты сюжета и развитие персонажей.
- [Консультант по путешествиям во времени](https://docs.claude.com/ru/resources/prompt-library/time-travel-consultant.md): Помогите пользователю разобраться в гипотетических сценариях путешествий во времени и их последствиях.
- [Скороговорка](https://docs.claude.com/ru/resources/prompt-library/tongue-twister.md): Создавайте сложные скороговорки.
- [Генератор викторин](https://docs.claude.com/ru/resources/prompt-library/trivia-generator.md): Создавайте вопросы для викторин на широкий спектр тем и предоставляйте подсказки при необходимости.
- [Детектор тона твитов](https://docs.claude.com/ru/resources/prompt-library/tweet-tone-detector.md): Определяйте тон и настроение твитов.
- [Инноватор VR фитнеса](https://docs.claude.com/ru/resources/prompt-library/vr-fitness-innovator.md): Генерируйте креативные идеи для игр виртуальной реальности в области фитнеса.
- [Мастер создания веб-сайтов](https://docs.claude.com/ru/resources/prompt-library/website-wizard.md): Создавайте одностраничные веб-сайты на основе пользовательских спецификаций.
- [Быстрые старты](https://docs.claude.com/ru/resources/quickstarts.md): Быстро начните работу с Claude и Claude API
- [Статус системы](https://docs.claude.com/ru/resources/status.md)
- [获取 API 密钥](https://docs.claude.com/zh-CN/api/admin-api/apikeys/get-api-key.md)
- [列出 API 密钥](https://docs.claude.com/zh-CN/api/admin-api/apikeys/list-api-keys.md)
- [更新 API 密钥](https://docs.claude.com/zh-CN/api/admin-api/apikeys/update-api-key.md)
- [获取 Claude Code 使用报告](https://docs.claude.com/zh-CN/api/admin-api/claude-code/get-claude-code-usage-report.md): Retrieve daily aggregated usage metrics for Claude Code users.
Enables organizations to analyze developer productivity and build custom dashboards.
- [创建邀请](https://docs.claude.com/zh-CN/api/admin-api/invites/create-invite.md)
- [删除邀请](https://docs.claude.com/zh-CN/api/admin-api/invites/delete-invite.md)
- [获取邀请](https://docs.claude.com/zh-CN/api/admin-api/invites/get-invite.md)
- [列出邀请](https://docs.claude.com/zh-CN/api/admin-api/invites/list-invites.md)
- [获取组织信息](https://docs.claude.com/zh-CN/api/admin-api/organization/get-me.md)
- [获取成本报告](https://docs.claude.com/zh-CN/api/admin-api/usage-cost/get-cost-report.md)
- [获取消息 API 的使用报告](https://docs.claude.com/zh-CN/api/admin-api/usage-cost/get-messages-usage-report.md)
- [获取用户](https://docs.claude.com/zh-CN/api/admin-api/users/get-user.md)
- [列出用户](https://docs.claude.com/zh-CN/api/admin-api/users/list-users.md)
- [移除用户](https://docs.claude.com/zh-CN/api/admin-api/users/remove-user.md)
- [更新用户](https://docs.claude.com/zh-CN/api/admin-api/users/update-user.md)
- [添加工作区成员](https://docs.claude.com/zh-CN/api/admin-api/workspace_members/create-workspace-member.md)
- [删除工作空间成员](https://docs.claude.com/zh-CN/api/admin-api/workspace_members/delete-workspace-member.md)
- [获取工作区成员](https://docs.claude.com/zh-CN/api/admin-api/workspace_members/get-workspace-member.md)
- [列出工作空间成员](https://docs.claude.com/zh-CN/api/admin-api/workspace_members/list-workspace-members.md)
- [更新工作区成员](https://docs.claude.com/zh-CN/api/admin-api/workspace_members/update-workspace-member.md)
- [归档工作区](https://docs.claude.com/zh-CN/api/admin-api/workspaces/archive-workspace.md)
- [创建工作区](https://docs.claude.com/zh-CN/api/admin-api/workspaces/create-workspace.md)
- [获取工作区](https://docs.claude.com/zh-CN/api/admin-api/workspaces/get-workspace.md)
- [列出工作区](https://docs.claude.com/zh-CN/api/admin-api/workspaces/list-workspaces.md)
- [更新工作区](https://docs.claude.com/zh-CN/api/admin-api/workspaces/update-workspace.md)
- [Admin API 概述](https://docs.claude.com/zh-CN/api/administration-api.md)
- [Beta 头部](https://docs.claude.com/zh-CN/api/beta-headers.md): 使用 Claude API 的 beta 头部的文档
- [取消消息批次](https://docs.claude.com/zh-CN/api/canceling-message-batches.md): Batches may be canceled any time before processing ends. Once cancellation is initiated, the batch enters a `canceling` state, at which time the system may complete any in-progress, non-interruptible requests before finalizing cancellation.
The number of canceled requests is specified in `request_counts`. To determine which requests were canceled, check the individual results within the batch. Note that cancellation may not result in any canceled requests if they were non-interruptible.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Claude Code Analytics API](https://docs.claude.com/zh-CN/api/claude-code-analytics-api.md): 通过 Claude Code Analytics Admin API 以编程方式访问您组织的 Claude Code 使用分析和生产力指标。
- [Amazon Bedrock API](https://docs.claude.com/zh-CN/api/claude-on-amazon-bedrock.md): Anthropic的Claude模型现已通过Amazon Bedrock正式可用。
- [Vertex AI API](https://docs.claude.com/zh-CN/api/claude-on-vertex-ai.md): Anthropic的Claude模型现已通过[Vertex AI](https://cloud.google.com/vertex-ai)正式提供。
- [客户端 SDK](https://docs.claude.com/zh-CN/api/client-sdks.md): 我们提供多种流行语言的客户端库，让您更轻松地使用 Claude API。
- [创建消息批处理](https://docs.claude.com/zh-CN/api/creating-message-batches.md): Send a batch of Message creation requests.
The Message Batches API can be used to process multiple Messages API requests at once. Once a Message Batch is created, it begins processing immediately. Batches can take up to 24 hours to complete.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [删除消息批次](https://docs.claude.com/zh-CN/api/deleting-message-batches.md): Delete a Message Batch.
Message Batches can only be deleted once they've finished processing. If you'd like to delete an in-progress batch, you must first cancel it.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [错误](https://docs.claude.com/zh-CN/api/errors.md)
- [下载文件](https://docs.claude.com/zh-CN/api/files-content.md): 下载Claude生成文件的内容
- [创建文件](https://docs.claude.com/zh-CN/api/files-create.md): 上传文件
- [删除文件](https://docs.claude.com/zh-CN/api/files-delete.md): 通过API使文件无法访问
- [列出文件](https://docs.claude.com/zh-CN/api/files-list.md): 列出工作区内的文件
- [获取文件元数据](https://docs.claude.com/zh-CN/api/files-metadata.md)
- [获取帮助](https://docs.claude.com/zh-CN/api/getting-help.md): 我们已经尝试在这些文档中提供最常见问题的答案。但是，如果您在使用 Claude、Claude API 或我们的任何产品时需要进一步的技术支持，您可以通过 [support.claude.com](https://support.claude.com) 联系我们的支持团队。
- [处理停止原因](https://docs.claude.com/zh-CN/api/handling-stop-reasons.md): 了解如何处理Claude API响应中的stop_reason字段，包括不同停止原因的含义和最佳实践。
- [IP 地址](https://docs.claude.com/zh-CN/api/ip-addresses.md): Anthropic 服务对入站和出站连接都使用固定的 IP 地址。您可以使用这些地址来配置防火墙规则，以安全访问 Claude API 和控制台。这些地址不会在没有通知的情况下更改。
- [列出消息批次](https://docs.claude.com/zh-CN/api/listing-message-batches.md): List all Message Batches within a Workspace. Most recently created batches are returned first.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [消息](https://docs.claude.com/zh-CN/api/messages.md): Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless multi-turn conversations.
Learn more about the Messages API in our [user guide](/en/docs/initial-setup)
- [Message Batches 示例](https://docs.claude.com/zh-CN/api/messages-batch-examples.md): Message Batches API 的示例用法
- [计算消息令牌数](https://docs.claude.com/zh-CN/api/messages-count-tokens.md): Count the number of tokens in a Message.
The Token Count API can be used to count the number of tokens in a Message, including tools, images, and documents, without creating it.
Learn more about token counting in our [user guide](/en/docs/build-with-claude/token-counting)
- [Messages 示例](https://docs.claude.com/zh-CN/api/messages-examples.md): Messages API 的请求和响应示例
- [从文本补全迁移](https://docs.claude.com/zh-CN/api/migrating-from-text-completions-to-messages.md): 从文本补全迁移到消息
- [获取模型](https://docs.claude.com/zh-CN/api/models.md): Get a specific model.
The Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.
- [列出模型](https://docs.claude.com/zh-CN/api/models-list.md): List available models.
The Models API response can be used to determine which models are available for use in the API. More recently released models are listed first.
- [OpenAI SDK 兼容性](https://docs.claude.com/zh-CN/api/openai-sdk.md): Anthropic 提供了一个兼容层，使您能够使用 OpenAI SDK 来测试 Claude API。通过少量代码更改，您可以快速评估 Anthropic 模型的能力。
- [概述](https://docs.claude.com/zh-CN/api/overview.md): Claude API 概述，包括访问方式、身份验证、内容类型、请求大小限制和响应头信息。
- [生成提示词](https://docs.claude.com/zh-CN/api/prompt-tools-generate.md): 生成一个写得很好的提示词
- [改进提示词](https://docs.claude.com/zh-CN/api/prompt-tools-improve.md): 在反馈指导下创建新的改进提示词
- [模板化提示词](https://docs.claude.com/zh-CN/api/prompt-tools-templatize.md): 通过识别和提取变量来模板化提示词
- [速率限制](https://docs.claude.com/zh-CN/api/rate-limits.md): 为了减少滥用并管理我们API的容量，我们对组织使用Claude API的程度实施了限制。
- [检索消息批次结果](https://docs.claude.com/zh-CN/api/retrieving-message-batch-results.md): Streams the results of a Message Batch as a `.jsonl` file.
Each line in the file is a JSON object containing the result of a single request in the Message Batch. Results are not guaranteed to be in the same order as requests. Use the `custom_id` field to match results to requests.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [检索消息批次](https://docs.claude.com/zh-CN/api/retrieving-message-batches.md): This endpoint is idempotent and can be used to poll for Message Batch completion. To access the results of a Message Batch, make a request to the `results_url` field in the response.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [服务层级](https://docs.claude.com/zh-CN/api/service-tiers.md): 不同的服务层级让您能够根据应用程序的需求平衡可用性、性能和可预测的成本。
- [支持的地区](https://docs.claude.com/zh-CN/api/supported-regions.md): 以下是我们目前可以支持访问的国家、地区和领土：
- [使用量和成本 API](https://docs.claude.com/zh-CN/api/usage-cost-api.md): 通过使用量和成本管理 API 以编程方式访问您组织的 API 使用量和成本数据。
- [版本](https://docs.claude.com/zh-CN/api/versioning.md): 在发出 API 请求时，您必须发送一个 `anthropic-version` 请求头。例如，`anthropic-version: 2023-06-01`。如果您使用我们的[客户端 SDK](/zh-CN/api/client-sdks)，这将自动为您处理。
- [术语表](https://docs.claude.com/zh-CN/docs/about-claude/glossary.md): 这些概念并非Anthropic语言模型独有，但我们在下面简要总结了关键术语。
- [模型弃用](https://docs.claude.com/zh-CN/docs/about-claude/model-deprecations.md): 了解Anthropic模型的弃用时间表、迁移指南和最佳实践。
- [选择合适的模型](https://docs.claude.com/zh-CN/docs/about-claude/models/choosing-a-model.md): 为您的应用选择最佳Claude模型需要平衡三个关键考虑因素：能力、速度和成本。本指南帮助您根据特定需求做出明智的决定。
- [迁移到 Claude 4](https://docs.claude.com/zh-CN/docs/about-claude/models/migrating-to-claude-4.md): 本页面提供从 Claude 3.7 模型迁移到 Claude 4 模型的指导。
- [模型概览](https://docs.claude.com/zh-CN/docs/about-claude/models/overview.md): Claude是由Anthropic开发的最先进的大型语言模型系列。本指南介绍我们的模型并比较它们与传统模型的性能。
- [定价](https://docs.claude.com/zh-CN/docs/about-claude/pricing.md): 了解 Anthropic 的模型和功能定价结构
- [内容审核](https://docs.claude.com/zh-CN/docs/about-claude/use-case-guides/content-moderation.md): 内容审核是维护数字应用程序中安全、尊重和高效环境的关键方面。在本指南中，我们将讨论如何使用 Claude 在您的数字应用程序中审核内容。
- [客户支持代理](https://docs.claude.com/zh-CN/docs/about-claude/use-case-guides/customer-support-chat.md): 本指南介绍如何利用Claude的高级对话能力来实时处理客户咨询，提供24/7支持，减少等待时间，并通过准确的回复和积极的互动来管理大量支持请求。
- [法律文件摘要](https://docs.claude.com/zh-CN/docs/about-claude/use-case-guides/legal-summarization.md): 本指南介绍如何利用Claude的高级自然语言处理能力高效地总结法律文件，提取关键信息并加速法律研究。通过Claude，您可以简化合同审查、诉讼准备和监管工作，节省时间并确保法律流程的准确性。
- [常见用例指南](https://docs.claude.com/zh-CN/docs/about-claude/use-case-guides/overview.md)
- [工单路由](https://docs.claude.com/zh-CN/docs/about-claude/use-case-guides/ticket-routing.md): 本指南介绍如何利用 Claude 的高级自然语言理解能力，基于客户意图、紧急程度、优先级、客户档案等因素大规模分类客户支持工单。
- [Google Sheets 插件](https://docs.claude.com/zh-CN/docs/agents-and-tools/claude-for-sheets.md): [Claude for Sheets 扩展](https://workspace.google.com/marketplace/app/claude%5Ffor%5Fsheets/909417792257)将 Claude 集成到 Google Sheets 中，允许您直接在单元格中执行与 Claude 的交互。
- [MCP 连接器](https://docs.claude.com/zh-CN/docs/agents-and-tools/mcp-connector.md): Claude 的模型上下文协议 (MCP) 连接器功能使您能够直接从 Messages API 连接到远程 MCP 服务器，无需单独的 MCP 客户端。
- [远程 MCP 服务器](https://docs.claude.com/zh-CN/docs/agents-and-tools/remote-mcp-servers.md): 了解如何连接到第三方部署的远程 MCP 服务器，扩展 Claude API 的功能。
- [Bash 工具](https://docs.claude.com/zh-CN/docs/agents-and-tools/tool-use/bash-tool.md): Bash 工具使 Claude 能够在持久的 bash 会话中执行 shell 命令，允许系统操作、脚本执行和命令行自动化。
- [代码执行工具](https://docs.claude.com/zh-CN/docs/agents-and-tools/tool-use/code-execution-tool.md): Claude可以分析数据、创建可视化、执行复杂计算、运行系统命令、创建和编辑文件，并直接在API对话中处理上传的文件。
- [计算机使用工具](https://docs.claude.com/zh-CN/docs/agents-and-tools/tool-use/computer-use-tool.md): Claude 可以通过计算机使用工具与计算机环境交互，该工具提供截图功能和鼠标/键盘控制，实现自主桌面交互。
- [细粒度工具流式传输](https://docs.claude.com/zh-CN/docs/agents-and-tools/tool-use/fine-grained-tool-streaming.md): 细粒度工具流式传输
- [如何实现工具使用](https://docs.claude.com/zh-CN/docs/agents-and-tools/tool-use/implement-tool-use.md): 学习如何在Claude中实现工具使用功能，包括模型选择、工具定义、控制输出和处理错误的完整指南。
- [Claude 工具使用](https://docs.claude.com/zh-CN/docs/agents-and-tools/tool-use/overview.md)
- [文本编辑器工具](https://docs.claude.com/zh-CN/docs/agents-and-tools/tool-use/text-editor-tool.md): 使用 Anthropic 定义的文本编辑器工具让 Claude 查看和修改文本文件，帮助您调试、修复和改进代码或其他文本文档。
- [高效令牌工具使用](https://docs.claude.com/zh-CN/docs/agents-and-tools/tool-use/token-efficient-tool-use.md): 了解如何使用Claude Sonnet 3.7的高效令牌工具使用功能来减少令牌消耗和延迟。
- [网页获取工具](https://docs.claude.com/zh-CN/docs/agents-and-tools/tool-use/web-fetch-tool.md)
- [网络搜索工具](https://docs.claude.com/zh-CN/docs/agents-and-tools/tool-use/web-search-tool.md): 网络搜索工具为Claude提供对实时网络内容的直接访问，使其能够使用超出其知识截止日期的最新信息回答问题。
- [批处理](https://docs.claude.com/zh-CN/docs/build-with-claude/batch-processing.md): 批处理是一种高效处理大量请求的强大方法。
- [引用](https://docs.claude.com/zh-CN/docs/build-with-claude/citations.md): Claude能够在回答有关文档的问题时提供详细的引用，帮助您跟踪和验证响应中的信息来源。
- [上下文窗口](https://docs.claude.com/zh-CN/docs/build-with-claude/context-windows.md): 了解Claude模型的上下文窗口如何工作，包括扩展思考和工具使用的特殊行为。
- [嵌入向量](https://docs.claude.com/zh-CN/docs/build-with-claude/embeddings.md): 文本嵌入向量是文本的数值表示，能够测量语义相似性。本指南介绍嵌入向量、其应用以及如何使用嵌入模型进行搜索、推荐和异常检测等任务。
- [使用扩展思维构建](https://docs.claude.com/zh-CN/docs/build-with-claude/extended-thinking.md)
- [Files API](https://docs.claude.com/zh-CN/docs/build-with-claude/files.md): Files API 让您上传和管理文件以与 Claude API 一起使用，无需在每次请求时重新上传内容。
- [多语言支持](https://docs.claude.com/zh-CN/docs/build-with-claude/multilingual-support.md): Claude在多种语言的任务中表现出色，相对于英语保持强大的跨语言性能。
- [功能概览](https://docs.claude.com/zh-CN/docs/build-with-claude/overview.md): 探索Claude的高级功能和能力。
- [PDF 支持](https://docs.claude.com/zh-CN/docs/build-with-claude/pdf-support.md): 使用 Claude 处理 PDF。从文档中提取文本、分析图表并理解视觉内容。
- [提示缓存](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-caching.md)
- [保持清晰、直接和详细](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/be-clear-and-direct.md)
- [让 Claude 思考（思维链提示）以提高性能](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/chain-of-thought.md)
- [链式复杂提示以获得更强的性能](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/chain-prompts.md)
- [Claude 4 提示工程最佳实践](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/claude-4-best-practices.md): Claude 4 模型的特定提示工程技术指南，帮助您在应用程序中获得最佳结果。
- [扩展思考技巧](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/extended-thinking-tips.md): 获得Claude扩展思考功能最佳效果的高级策略和技术指南。
- [长文本提示技巧](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/long-context-tips.md)
- [使用示例（多示例提示）来引导 Claude 的行为](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/multishot-prompting.md)
- [提示工程概述](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/overview.md)
- [预填充 Claude 的回复以获得更好的输出控制](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/prefill-claudes-response.md)
- [自动生成首版提示词模板](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/prompt-generator.md)
- [使用我们的提示词改进器来优化您的提示词](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/prompt-improver.md): 使用我们的提示词改进器来优化您的提示词
- [使用提示模板和变量](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/prompt-templates-and-variables.md): 学习如何在Claude中使用提示模板和变量来创建更高效、一致和可扩展的AI应用程序。
- [通过系统提示给Claude赋予角色](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/system-prompts.md)
- [使用XML标签构建你的提示词](https://docs.claude.com/zh-CN/docs/build-with-claude/prompt-engineering/use-xml-tags.md)
- [搜索结果](https://docs.claude.com/zh-CN/docs/build-with-claude/search-results.md): 通过提供带有来源归属的搜索结果，为RAG应用程序启用自然引用
- [流式消息](https://docs.claude.com/zh-CN/docs/build-with-claude/streaming.md): 了解如何使用服务器发送事件(SSE)增量流式传输Claude的响应。
- [Token 计数](https://docs.claude.com/zh-CN/docs/build-with-claude/token-counting.md): Token 计数使您能够在发送消息给 Claude 之前确定消息中的 token 数量，帮助您对提示和使用做出明智的决策。
- [视觉](https://docs.claude.com/zh-CN/docs/build-with-claude/vision.md): Claude 3 和 4 系列模型具备新的视觉功能，允许 Claude 理解和分析图像，为多模态交互开启了令人兴奋的可能性。
- [Amazon Bedrock 上的 Claude Code](https://docs.claude.com/zh-CN/docs/claude-code/amazon-bedrock.md): 了解如何通过 Amazon Bedrock 配置 Claude Code，包括设置、IAM 配置和故障排除。
- [分析](https://docs.claude.com/zh-CN/docs/claude-code/analytics.md): 查看您组织的 Claude Code 部署的详细使用洞察和生产力指标。
- [CLI 参考](https://docs.claude.com/zh-CN/docs/claude-code/cli-reference.md): Claude Code 命令行界面的完整参考，包括命令和标志。
- [常见工作流程](https://docs.claude.com/zh-CN/docs/claude-code/common-workflows.md): 了解使用 Claude Code 的常见工作流程。
- [有效管理成本](https://docs.claude.com/zh-CN/docs/claude-code/costs.md): 了解如何在使用 Claude Code 时跟踪和优化令牌使用量和成本。
- [数据使用](https://docs.claude.com/zh-CN/docs/claude-code/data-usage.md): 了解Anthropic关于Claude的数据使用政策
- [开发容器](https://docs.claude.com/zh-CN/docs/claude-code/devcontainer.md): 了解 Claude Code 开发容器，适用于需要一致、安全环境的团队。
- [Claude Code GitHub Actions](https://docs.claude.com/zh-CN/docs/claude-code/github-actions.md): 了解如何使用 Claude Code GitHub Actions 将 Claude Code 集成到您的开发工作流程中
- [Claude Code GitLab CI/CD](https://docs.claude.com/zh-CN/docs/claude-code/gitlab-ci-cd.md): 了解如何将 Claude Code 集成到您的 GitLab CI/CD 开发工作流程中
- [Google Vertex AI 上的 Claude Code](https://docs.claude.com/zh-CN/docs/claude-code/google-vertex-ai.md): 了解如何通过 Google Vertex AI 配置 Claude Code，包括设置、IAM 配置和故障排除。
- [Hooks 参考](https://docs.claude.com/zh-CN/docs/claude-code/hooks.md): 本页面提供在 Claude Code 中实现 hooks 的参考文档。
- [Claude Code hooks 入门指南](https://docs.claude.com/zh-CN/docs/claude-code/hooks-guide.md): 学习如何通过注册 shell 命令来自定义和扩展 Claude Code 的行为
- [身份和访问管理](https://docs.claude.com/zh-CN/docs/claude-code/iam.md): 了解如何为您的组织配置Claude Code的用户身份验证、授权和访问控制。
- [将 Claude Code 添加到您的 IDE](https://docs.claude.com/zh-CN/docs/claude-code/ide-integrations.md): 了解如何将 Claude Code 添加到您喜爱的 IDE
- [交互模式](https://docs.claude.com/zh-CN/docs/claude-code/interactive-mode.md): Claude Code 会话中键盘快捷键、输入模式和交互功能的完整参考。
- [法律和合规](https://docs.claude.com/zh-CN/docs/claude-code/legal-and-compliance.md): Claude Code的法律协议、合规认证和安全信息。
- [LLM 网关配置](https://docs.claude.com/zh-CN/docs/claude-code/llm-gateway.md): 了解如何使用 LLM 网关解决方案配置 Claude Code，包括 LiteLLM 设置、身份验证方法以及使用跟踪和预算管理等企业功能。
- [通过 MCP 将 Claude Code 连接到工具](https://docs.claude.com/zh-CN/docs/claude-code/mcp.md): 了解如何使用 Model Context Protocol 将 Claude Code 连接到您的工具。
- [管理 Claude 的内存](https://docs.claude.com/zh-CN/docs/claude-code/memory.md): 学习如何通过不同的内存位置和最佳实践来管理 Claude Code 跨会话的内存。
- [模型配置](https://docs.claude.com/zh-CN/docs/claude-code/model-config.md): 了解 Claude Code 模型配置，包括像 `opusplan` 这样的模型别名
- [监控](https://docs.claude.com/zh-CN/docs/claude-code/monitoring-usage.md): 了解如何为 Claude Code 启用和配置 OpenTelemetry。
- [企业网络配置](https://docs.claude.com/zh-CN/docs/claude-code/network-config.md): 为企业环境配置 Claude Code，支持代理服务器、自定义证书颁发机构 (CA) 和双向传输层安全 (mTLS) 身份验证。
- [输出样式](https://docs.claude.com/zh-CN/docs/claude-code/output-styles.md): 将 Claude Code 适配用于软件工程之外的用途
- [Claude Code 概述](https://docs.claude.com/zh-CN/docs/claude-code/overview.md): 了解 Claude Code，Anthropic 的智能编程工具，它运行在您的终端中，帮助您比以往任何时候都更快地将想法转化为代码。
- [快速开始](https://docs.claude.com/zh-CN/docs/claude-code/quickstart.md): 欢迎使用 Claude Code！
- [自定义工具](https://docs.claude.com/zh-CN/docs/claude-code/sdk/custom-tools.md): 构建和集成自定义工具以扩展Claude Code SDK功能
- [修改系统提示](https://docs.claude.com/zh-CN/docs/claude-code/sdk/modifying-system-prompts.md): 学习如何通过三种方法修改系统提示来自定义Claude的行为 - 输出样式、appendSystemPrompt和customSystemPrompt。
- [跟踪成本和使用情况](https://docs.claude.com/zh-CN/docs/claude-code/sdk/sdk-cost-tracking.md): 了解并跟踪 Claude Code SDK 中用于计费的令牌使用情况
- [无头模式](https://docs.claude.com/zh-CN/docs/claude-code/sdk/sdk-headless.md): 以编程方式运行 Claude Code，无需交互式 UI
- [SDK 中的 MCP](https://docs.claude.com/zh-CN/docs/claude-code/sdk/sdk-mcp.md): 使用模型上下文协议服务器通过自定义工具扩展 Claude Code
- [概述](https://docs.claude.com/zh-CN/docs/claude-code/sdk/sdk-overview.md): 使用 Claude Code SDK 构建自定义 AI 智能体
- [处理权限](https://docs.claude.com/zh-CN/docs/claude-code/sdk/sdk-permissions.md): 在 Claude Code SDK 中控制工具使用和权限
- [Python SDK 参考](https://docs.claude.com/zh-CN/docs/claude-code/sdk/sdk-python.md): Claude Code Python SDK 的完整 API 参考，包括所有函数、类型和类。
- [会话管理](https://docs.claude.com/zh-CN/docs/claude-code/sdk/sdk-sessions.md): 了解 Claude Code SDK 如何处理会话和会话恢复
- [SDK 中的斜杠命令](https://docs.claude.com/zh-CN/docs/claude-code/sdk/sdk-slash-commands.md): 学习如何使用斜杠命令通过 SDK 控制 Claude Code 会话
- [TypeScript SDK 参考](https://docs.claude.com/zh-CN/docs/claude-code/sdk/sdk-typescript.md): Claude Code TypeScript SDK 的完整 API 参考，包括所有函数、类型和接口。
- [流式输入](https://docs.claude.com/zh-CN/docs/claude-code/sdk/streaming-vs-single-mode.md): 了解 Claude Code SDK 的两种输入模式以及何时使用每种模式
- [SDK 中的子代理](https://docs.claude.com/zh-CN/docs/claude-code/sdk/subagents.md): 在 Claude Code SDK 中使用子代理
- [待办事项列表](https://docs.claude.com/zh-CN/docs/claude-code/sdk/todo-tracking.md): 使用 Claude Code SDK 跟踪和显示待办事项，实现有组织的任务管理
- [安全性](https://docs.claude.com/zh-CN/docs/claude-code/security.md): 了解 Claude Code 的安全保障措施和安全使用的最佳实践。
- [Claude Code 设置](https://docs.claude.com/zh-CN/docs/claude-code/settings.md): 使用全局和项目级设置以及环境变量配置 Claude Code。
- [设置 Claude Code](https://docs.claude.com/zh-CN/docs/claude-code/setup.md): 在您的开发机器上安装、认证并开始使用 Claude Code。
- [斜杠命令](https://docs.claude.com/zh-CN/docs/claude-code/slash-commands.md): 在交互式会话中使用斜杠命令控制 Claude 的行为。
- [状态栏配置](https://docs.claude.com/zh-CN/docs/claude-code/statusline.md): 为 Claude Code 创建自定义状态栏以显示上下文信息
- [子代理](https://docs.claude.com/zh-CN/docs/claude-code/sub-agents.md): 在Claude Code中创建和使用专门的AI子代理，用于特定任务的工作流程和改进的上下文管理。
- [优化您的终端设置](https://docs.claude.com/zh-CN/docs/claude-code/terminal-config.md): 当您的终端配置正确时，Claude Code 工作效果最佳。遵循这些指南来优化您的体验。
- [企业部署概述](https://docs.claude.com/zh-CN/docs/claude-code/third-party-integrations.md): 了解Claude Code如何与各种第三方服务和基础设施集成，以满足企业部署要求。
- [故障排除](https://docs.claude.com/zh-CN/docs/claude-code/troubleshooting.md): 发现Claude Code安装和使用中常见问题的解决方案。
- [开始使用 Claude](https://docs.claude.com/zh-CN/docs/get-started.md): 进行您的第一次 Claude API 调用并构建一个简单的网络搜索助手
- [Claude 简介](https://docs.claude.com/zh-CN/docs/intro.md): Claude 是由 Anthropic 构建的高性能、可信赖且智能的 AI 平台。Claude 在涉及语言、推理、分析、编程等任务方面表现出色。
- [Anthropic 隐私政策](https://docs.claude.com/zh-CN/docs/legal-center/privacy.md)
- [安全与合规](https://docs.claude.com/zh-CN/docs/legal-center/security-compliance.md)
- [使用 Claude 构建](https://docs.claude.com/zh-CN/docs/overview.md): Claude 是由 Anthropic 构建的一系列[高性能和智能的 AI 模型](/zh-CN/docs/about-claude/models)。虽然 Claude 功能强大且可扩展，但它也是最值得信赖和可靠的 AI。它遵循关键协议，犯错更少，并且抗越狱攻击——让[企业客户](https://claude.com/customers)能够大规模构建最安全的 AI 驱动应用程序。
- [定义你的成功标准](https://docs.claude.com/zh-CN/docs/test-and-evaluate/define-success.md)
- [创建强有力的实证评估](https://docs.claude.com/zh-CN/docs/test-and-evaluate/develop-tests.md): 学习如何开发测试用例来衡量LLM性能，包括评估设计原则、示例评估和评分方法。
- [使用评估工具](https://docs.claude.com/zh-CN/docs/test-and-evaluate/eval-tool.md): [Claude Console](https://console.anthropic.com/dashboard) 提供了一个**评估工具**，允许您在各种场景下测试您的提示词。
- [null](https://docs.claude.com/zh-CN/docs/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals.md)
- [提高输出一致性（JSON模式）](https://docs.claude.com/zh-CN/docs/test-and-evaluate/strengthen-guardrails/increase-consistency.md)
- [通过角色提示和预填充保持Claude的角色特征](https://docs.claude.com/zh-CN/docs/test-and-evaluate/strengthen-guardrails/keep-claude-in-character.md)
- [缓解越狱和提示注入](https://docs.claude.com/zh-CN/docs/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks.md)
- [减少幻觉](https://docs.claude.com/zh-CN/docs/test-and-evaluate/strengthen-guardrails/reduce-hallucinations.md)
- [降低延迟](https://docs.claude.com/zh-CN/docs/test-and-evaluate/strengthen-guardrails/reduce-latency.md)
- [减少提示词泄露](https://docs.claude.com/zh-CN/docs/test-and-evaluate/strengthen-guardrails/reduce-prompt-leak.md)
- [null](https://docs.claude.com/zh-CN/home.md)
- [Claude 开发者平台](https://docs.claude.com/zh-CN/release-notes/api.md): 跟进 Claude API 和 Claude Console 的更新。
- [Claude Apps](https://docs.claude.com/zh-CN/release-notes/claude-apps.md): 跟进 Anthropic 的 Claude 应用程序的更新。
- [Claude Code](https://docs.claude.com/zh-CN/release-notes/claude-code.md): 了解 Claude Code 的最新功能和改进
- [概览](https://docs.claude.com/zh-CN/release-notes/overview.md): 跟进 Anthropic 产品和服务的更新。
- [系统提示词](https://docs.claude.com/zh-CN/release-notes/system-prompts.md): 查看 [Claude.ai](https://www.claude.ai) 和 Claude [iOS](http://anthropic.com/ios) 及 [Android](http://anthropic.com/android) 应用程序上核心系统提示词的更新。
- [Claude 3.7 系统卡](https://docs.claude.com/zh-CN/resources/claude-3-7-system-card.md)
- [Claude 3 模型卡](https://docs.claude.com/zh-CN/resources/claude-3-model-card.md)
- [Claude 4 系统卡片](https://docs.claude.com/zh-CN/resources/claude-4-system-card.md): Anthropic 的 Claude Opus 和 Sonnet 4 模型卡片，包含性能和安全详细信息。
- [Anthropic 课程](https://docs.claude.com/zh-CN/resources/courses.md)
- [null](https://docs.claude.com/zh-CN/resources/overview.md)
- [自适应编辑器](https://docs.claude.com/zh-CN/resources/prompt-library/adaptive-editor.md): 根据用户给出的指令重写文本，比如使用不同的语调、受众或风格。
- [机场代码分析器](https://docs.claude.com/zh-CN/resources/prompt-library/airport-code-analyst.md): 从文本中查找和提取机场代码。
- [外星人类学家](https://docs.claude.com/zh-CN/resources/prompt-library/alien-anthropologist.md): 从外星人类学家的视角分析人类文化和习俗。
- [头韵炼金术师](https://docs.claude.com/zh-CN/resources/prompt-library/alliteration-alchemist.md): 为任何给定主题生成头韵短语和句子。
- [Babel的广播](https://docs.claude.com/zh-CN/resources/prompt-library/babels-broadcasts.md): 用世界上10种使用最广泛的语言创建引人注目的产品发布推文。
- [品牌建设者](https://docs.claude.com/zh-CN/resources/prompt-library/brand-builder.md): 为整体品牌形象制作设计简报。
- [职业教练](https://docs.claude.com/zh-CN/resources/prompt-library/career-coach.md): 与AI职业教练进行角色扮演对话。
- [引用你的来源](https://docs.claude.com/zh-CN/resources/prompt-library/cite-your-sources.md): 获取关于文档内容问题的答案，并提供支持回答的相关引用。
- [代码解释器](https://docs.claude.com/zh-CN/resources/prompt-library/code-clarifier.md): 用简单的语言简化和解释复杂的代码。
- [代码顾问](https://docs.claude.com/zh-CN/resources/prompt-library/code-consultant.md): 建议改进以优化Python代码性能。
- [企业预言家](https://docs.claude.com/zh-CN/resources/prompt-library/corporate-clairvoyant.md): 从冗长的企业报告中提取洞察、识别风险并将关键信息提炼成单一备忘录。
- [宇宙按键](https://docs.claude.com/zh-CN/resources/prompt-library/cosmic-keystrokes.md): 在单个HTML文件中生成一个交互式速度打字游戏，具有横向滚动游戏玩法和Tailwind CSS样式。
- [CSV转换器](https://docs.claude.com/zh-CN/resources/prompt-library/csv-converter.md): 将各种格式（JSON、XML等）的数据转换为格式正确的CSV文件。
- [烹饪创作者](https://docs.claude.com/zh-CN/resources/prompt-library/culinary-creator.md): 根据用户的现有食材和饮食偏好建议食谱创意。
- [数据整理器](https://docs.claude.com/zh-CN/resources/prompt-library/data-organizer.md): 将非结构化文本转换为定制的JSON表格。
- [方向解码器](https://docs.claude.com/zh-CN/resources/prompt-library/direction-decoder.md): 将自然语言转换为逐步指导。
- [梦境解释师](https://docs.claude.com/zh-CN/resources/prompt-library/dream-interpreter.md): 为用户的梦境提供象征意义的解释和洞察。
- [效率估算器](https://docs.claude.com/zh-CN/resources/prompt-library/efficiency-estimator.md): 计算函数和算法的时间复杂度。
- [邮箱地址提取器](https://docs.claude.com/zh-CN/resources/prompt-library/email-extractor.md): 从文档中提取邮箱地址并生成JSON格式的列表。
- [表情符号编码器](https://docs.claude.com/zh-CN/resources/prompt-library/emoji-encoder.md): 将纯文本转换为有趣且富有表现力的表情符号消息。
- [道德困境导航器](https://docs.claude.com/zh-CN/resources/prompt-library/ethical-dilemma-navigator.md): 帮助用户思考复杂的道德困境并提供不同的视角。
- [Excel公式专家](https://docs.claude.com/zh-CN/resources/prompt-library/excel-formula-expert.md): 根据用户描述的计算或数据操作创建Excel公式。
- [函数制造器](https://docs.claude.com/zh-CN/resources/prompt-library/function-fabricator.md): 基于详细规范创建Python函数。
- [未来主义时尚顾问](https://docs.claude.com/zh-CN/resources/prompt-library/futuristic-fashion-advisor.md): 为用户的特定偏好建议前卫的时尚趋势和风格。
- [Git 精通](https://docs.claude.com/zh-CN/resources/prompt-library/git-gud.md): 根据用户描述的版本控制操作生成适当的 Git 命令。
- [Google apps 脚本编写器](https://docs.claude.com/zh-CN/resources/prompt-library/google-apps-scripter.md): 根据用户需求生成 Google Apps 脚本来完成任务。
- [评分专家](https://docs.claude.com/zh-CN/resources/prompt-library/grading-guru.md): 基于用户定义的标准和要求比较和评估书面文本的质量。
- [语法精灵](https://docs.claude.com/zh-CN/resources/prompt-library/grammar-genie.md): 将语法错误的句子转换为正确的英语。
- [Hal，幽默的助手](https://docs.claude.com/zh-CN/resources/prompt-library/hal-the-humorous-helper.md): 与一个知识渊博且带有讽刺色彩的AI聊天。
- [习语解释器](https://docs.claude.com/zh-CN/resources/prompt-library/idiom-illuminator.md): 解释常见习语和谚语的含义和起源。
- [面试问题制作器](https://docs.claude.com/zh-CN/resources/prompt-library/interview-question-crafter.md): 为面试生成问题。
- [LaTeX 传奇](https://docs.claude.com/zh-CN/resources/prompt-library/latex-legend.md): 编写 LaTeX 文档，生成数学方程式、表格等代码。
- [课程规划师](https://docs.claude.com/zh-CN/resources/prompt-library/lesson-planner.md): 制作任何学科的深入课程计划。
- [提示词库](https://docs.claude.com/zh-CN/resources/prompt-library/library.md)
- [主审核员](https://docs.claude.com/zh-CN/resources/prompt-library/master-moderator.md): 评估用户输入是否包含潜在有害或非法内容。
- [会议记录员](https://docs.claude.com/zh-CN/resources/prompt-library/meeting-scribe.md): 将会议内容提炼为简洁摘要，包括讨论主题、关键要点和行动项目。
- [备忘录大师](https://docs.claude.com/zh-CN/resources/prompt-library/memo-maestro.md): 基于关键要点撰写全面的公司备忘录。
- [正念导师](https://docs.claude.com/zh-CN/resources/prompt-library/mindfulness-mentor.md): 引导用户进行正念练习和减压技巧。
- [情绪着色器](https://docs.claude.com/zh-CN/resources/prompt-library/mood-colorizer.md): 将情绪的文本描述转换为相应的HEX颜色代码。
- [励志缪斯](https://docs.claude.com/zh-CN/resources/prompt-library/motivational-muse.md): 根据用户输入提供个性化的励志信息和肯定语句。
- [新词创造者](https://docs.claude.com/zh-CN/resources/prompt-library/neologism-creator.md): 基于用户提供的概念或想法发明新词并提供其定义。
- [观点思考者](https://docs.claude.com/zh-CN/resources/prompt-library/perspectives-ponderer.md): 权衡用户提供话题的利弊。
- [哲学思辨](https://docs.claude.com/zh-CN/resources/prompt-library/philosophical-musings.md): 参与深度哲学讨论和思想实验。
- [PII 净化器](https://docs.claude.com/zh-CN/resources/prompt-library/pii-purifier.md): 自动检测并从文本文档中删除个人身份信息（PII）。
- [多语言超能力](https://docs.claude.com/zh-CN/resources/prompt-library/polyglot-superpowers.md): 将任何语言的文本翻译成任何语言。
- [混成词诗人](https://docs.claude.com/zh-CN/resources/prompt-library/portmanteau-poet.md): 将两个词混合在一起，创造一个新的、有意义的混成词。
- [产品命名专家](https://docs.claude.com/zh-CN/resources/prompt-library/product-naming-pro.md): 根据描述和关键词创建吸引人的产品名称。
- [散文润色师](https://docs.claude.com/zh-CN/resources/prompt-library/prose-polisher.md): 使用高级文案编辑技术和建议来完善和改进书面内容。
- [双关语专家](https://docs.claude.com/zh-CN/resources/prompt-library/pun-dit.md): 基于任何给定主题生成巧妙的双关语和文字游戏。
- [Python 错误修复器](https://docs.claude.com/zh-CN/resources/prompt-library/python-bug-buster.md): 检测并修复 Python 代码中的错误。
- [评论分类器](https://docs.claude.com/zh-CN/resources/prompt-library/review-classifier.md): 将反馈分类到预先指定的标签和分类中。
- [猜猜这个谜语](https://docs.claude.com/zh-CN/resources/prompt-library/riddle-me-this.md): 生成谜语并引导用户找到解决方案。
- [科幻场景模拟器](https://docs.claude.com/zh-CN/resources/prompt-library/sci-fi-scenario-simulator.md): 与用户讨论各种科幻场景以及相关的挑战和考虑因素。
- [二年级简化器](https://docs.claude.com/zh-CN/resources/prompt-library/second-grade-simplifier.md): 让复杂文本变得易于年轻学习者理解。
- [比喻专家](https://docs.claude.com/zh-CN/resources/prompt-library/simile-savant.md): 从基本描述生成比喻。
- [苏格拉底式智者](https://docs.claude.com/zh-CN/resources/prompt-library/socratic-sage.md): 就用户给定的话题进行苏格拉底式对话。
- [电子表格魔法师](https://docs.claude.com/zh-CN/resources/prompt-library/spreadsheet-sorcerer.md): 生成包含各种类型数据的CSV电子表格。
- [SQL 魔法师](https://docs.claude.com/zh-CN/resources/prompt-library/sql-sorcerer.md): 将日常语言转换为 SQL 查询。
- [故事创作伙伴](https://docs.claude.com/zh-CN/resources/prompt-library/storytelling-sidekick.md): 与用户协作创造引人入胜的故事，提供情节转折和角色发展。
- [时间旅行顾问](https://docs.claude.com/zh-CN/resources/prompt-library/time-travel-consultant.md): 帮助用户探索假设的时间旅行场景及其影响。
- [绕口令](https://docs.claude.com/zh-CN/resources/prompt-library/tongue-twister.md): 创建具有挑战性的绕口令。
- [趣味问答生成器](https://docs.claude.com/zh-CN/resources/prompt-library/trivia-generator.md): 生成各种主题的趣味问答题目，并在需要时提供提示。
- [推文语调检测器](https://docs.claude.com/zh-CN/resources/prompt-library/tweet-tone-detector.md): 检测推文背后的语调和情感。
- [VR健身创新者](https://docs.claude.com/zh-CN/resources/prompt-library/vr-fitness-innovator.md): 为虚拟现实健身游戏头脑风暴创意想法。
- [网站向导](https://docs.claude.com/zh-CN/resources/prompt-library/website-wizard.md): 根据用户规格创建单页网站。
- [快速开始](https://docs.claude.com/zh-CN/resources/quickstarts.md): 快速开始使用 Claude 和 Claude API
- [系统状态](https://docs.claude.com/zh-CN/resources/status.md)
- [取得 API 金鑰](https://docs.claude.com/zh-TW/api/admin-api/apikeys/get-api-key.md)
- [列出 API 金鑰](https://docs.claude.com/zh-TW/api/admin-api/apikeys/list-api-keys.md)
- [更新 API 金鑰](https://docs.claude.com/zh-TW/api/admin-api/apikeys/update-api-key.md)
- [取得 Claude Code 使用報告](https://docs.claude.com/zh-TW/api/admin-api/claude-code/get-claude-code-usage-report.md): Retrieve daily aggregated usage metrics for Claude Code users.
Enables organizations to analyze developer productivity and build custom dashboards.
- [建立邀請](https://docs.claude.com/zh-TW/api/admin-api/invites/create-invite.md)
- [刪除邀請](https://docs.claude.com/zh-TW/api/admin-api/invites/delete-invite.md)
- [取得邀請](https://docs.claude.com/zh-TW/api/admin-api/invites/get-invite.md)
- [列出邀請](https://docs.claude.com/zh-TW/api/admin-api/invites/list-invites.md)
- [取得組織資訊](https://docs.claude.com/zh-TW/api/admin-api/organization/get-me.md)
- [取得成本報告](https://docs.claude.com/zh-TW/api/admin-api/usage-cost/get-cost-report.md)
- [取得 Messages API 的使用報告](https://docs.claude.com/zh-TW/api/admin-api/usage-cost/get-messages-usage-report.md)
- [取得使用者](https://docs.claude.com/zh-TW/api/admin-api/users/get-user.md)
- [列出使用者](https://docs.claude.com/zh-TW/api/admin-api/users/list-users.md)
- [移除使用者](https://docs.claude.com/zh-TW/api/admin-api/users/remove-user.md)
- [更新使用者](https://docs.claude.com/zh-TW/api/admin-api/users/update-user.md)
- [新增工作區成員](https://docs.claude.com/zh-TW/api/admin-api/workspace_members/create-workspace-member.md)
- [刪除工作區成員](https://docs.claude.com/zh-TW/api/admin-api/workspace_members/delete-workspace-member.md)
- [取得工作區成員](https://docs.claude.com/zh-TW/api/admin-api/workspace_members/get-workspace-member.md)
- [列出工作區成員](https://docs.claude.com/zh-TW/api/admin-api/workspace_members/list-workspace-members.md)
- [更新工作區成員](https://docs.claude.com/zh-TW/api/admin-api/workspace_members/update-workspace-member.md)
- [封存工作區](https://docs.claude.com/zh-TW/api/admin-api/workspaces/archive-workspace.md)
- [建立工作區](https://docs.claude.com/zh-TW/api/admin-api/workspaces/create-workspace.md)
- [取得工作區](https://docs.claude.com/zh-TW/api/admin-api/workspaces/get-workspace.md)
- [列出工作區](https://docs.claude.com/zh-TW/api/admin-api/workspaces/list-workspaces.md)
- [更新工作區](https://docs.claude.com/zh-TW/api/admin-api/workspaces/update-workspace.md)
- [Admin API 概述](https://docs.claude.com/zh-TW/api/administration-api.md)
- [Beta 標頭](https://docs.claude.com/zh-TW/api/beta-headers.md): 使用 Claude API 的 beta 標頭的文件
- [取消訊息批次](https://docs.claude.com/zh-TW/api/canceling-message-batches.md): Batches may be canceled any time before processing ends. Once cancellation is initiated, the batch enters a `canceling` state, at which time the system may complete any in-progress, non-interruptible requests before finalizing cancellation.
The number of canceled requests is specified in `request_counts`. To determine which requests were canceled, check the individual results within the batch. Note that cancellation may not result in any canceled requests if they were non-interruptible.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [Claude Code Analytics API](https://docs.claude.com/zh-TW/api/claude-code-analytics-api.md): 透過 Claude Code Analytics Admin API 程式化存取您組織的 Claude Code 使用分析和生產力指標。
- [Amazon Bedrock API](https://docs.claude.com/zh-TW/api/claude-on-amazon-bedrock.md): Anthropic 的 Claude 模型現已透過 Amazon Bedrock 正式提供。
- [Vertex AI API](https://docs.claude.com/zh-TW/api/claude-on-vertex-ai.md): Anthropic 的 Claude 模型現已透過 [Vertex AI](https://cloud.google.com/vertex-ai) 正式提供。
- [客戶端 SDK](https://docs.claude.com/zh-TW/api/client-sdks.md): 我們提供多種熱門程式語言的客戶端函式庫，讓您更輕鬆地使用 Claude API。
- [建立訊息批次](https://docs.claude.com/zh-TW/api/creating-message-batches.md): Send a batch of Message creation requests.
The Message Batches API can be used to process multiple Messages API requests at once. Once a Message Batch is created, it begins processing immediately. Batches can take up to 24 hours to complete.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [刪除訊息批次](https://docs.claude.com/zh-TW/api/deleting-message-batches.md): Delete a Message Batch.
Message Batches can only be deleted once they've finished processing. If you'd like to delete an in-progress batch, you must first cancel it.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [錯誤](https://docs.claude.com/zh-TW/api/errors.md): 了解 API 錯誤代碼、請求大小限制、錯誤格式和長時間請求的最佳實踐。
- [下載檔案](https://docs.claude.com/zh-TW/api/files-content.md): 下載 Claude 生成檔案的內容
- [建立檔案](https://docs.claude.com/zh-TW/api/files-create.md): 上傳檔案
- [刪除檔案](https://docs.claude.com/zh-TW/api/files-delete.md): 使檔案無法透過 API 存取
- [列出檔案](https://docs.claude.com/zh-TW/api/files-list.md): 列出工作區內的檔案
- [取得檔案中繼資料](https://docs.claude.com/zh-TW/api/files-metadata.md)
- [獲得幫助](https://docs.claude.com/zh-TW/api/getting-help.md): 我們已經嘗試在這些文件中提供最常見問題的答案。然而，如果您在使用 Claude、Claude API 或我們的任何產品時需要進一步的技術支援，您可以在 [support.claude.com](https://support.claude.com) 聯繫我們的支援團隊。
- [處理停止原因](https://docs.claude.com/zh-TW/api/handling-stop-reasons.md): 了解如何處理 Claude API 回應中的 stop_reason 欄位，以建立更強健的應用程式。
- [IP 地址](https://docs.claude.com/zh-TW/api/ip-addresses.md): Anthropic 服務對入站和出站連接都使用固定的 IP 地址。您可以使用這些地址來配置防火牆規則，以安全地存取 Claude API 和控制台。這些地址不會在未通知的情況下更改。
- [列出訊息批次](https://docs.claude.com/zh-TW/api/listing-message-batches.md): List all Message Batches within a Workspace. Most recently created batches are returned first.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [訊息](https://docs.claude.com/zh-TW/api/messages.md): Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless multi-turn conversations.
Learn more about the Messages API in our [user guide](/en/docs/initial-setup)
- [Message Batches 範例](https://docs.claude.com/zh-TW/api/messages-batch-examples.md): Message Batches API 的使用範例
- [計算訊息代幣](https://docs.claude.com/zh-TW/api/messages-count-tokens.md): Count the number of tokens in a Message.
The Token Count API can be used to count the number of tokens in a Message, including tools, images, and documents, without creating it.
Learn more about token counting in our [user guide](/en/docs/build-with-claude/token-counting)
- [Messages 範例](https://docs.claude.com/zh-TW/api/messages-examples.md): Messages API 的請求和回應範例
- [從文本補全遷移](https://docs.claude.com/zh-TW/api/migrating-from-text-completions-to-messages.md): 從文本補全遷移到訊息
- [取得模型](https://docs.claude.com/zh-TW/api/models.md): Get a specific model.
The Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.
- [列出模型](https://docs.claude.com/zh-TW/api/models-list.md): List available models.
The Models API response can be used to determine which models are available for use in the API. More recently released models are listed first.
- [OpenAI SDK 相容性](https://docs.claude.com/zh-TW/api/openai-sdk.md): Anthropic 提供了一個相容性層，讓您能夠使用 OpenAI SDK 來測試 Claude API。透過一些程式碼變更，您可以快速評估 Anthropic 模型的能力。
- [概述](https://docs.claude.com/zh-TW/api/overview.md): Claude API 的概述，包括存取方式、驗證、內容類型、請求大小限制、回應標頭和範例。
- [生成提示詞](https://docs.claude.com/zh-TW/api/prompt-tools-generate.md): 生成一個寫得很好的提示詞
- [改進提示](https://docs.claude.com/zh-TW/api/prompt-tools-improve.md): 根據回饋創建一個新的改進提示
- [模板化提示](https://docs.claude.com/zh-TW/api/prompt-tools-templatize.md): 通過識別和提取變數來模板化提示
- [速率限制](https://docs.claude.com/zh-TW/api/rate-limits.md): 為了減少濫用並管理我們 API 的容量，我們對組織使用 Claude API 的程度實施了限制。
- [檢索訊息批次結果](https://docs.claude.com/zh-TW/api/retrieving-message-batch-results.md): Streams the results of a Message Batch as a `.jsonl` file.
Each line in the file is a JSON object containing the result of a single request in the Message Batch. Results are not guaranteed to be in the same order as requests. Use the `custom_id` field to match results to requests.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [檢索訊息批次](https://docs.claude.com/zh-TW/api/retrieving-message-batches.md): This endpoint is idempotent and can be used to poll for Message Batch completion. To access the results of a Message Batch, make a request to the `results_url` field in the response.
Learn more about the Message Batches API in our [user guide](/en/docs/build-with-claude/batch-processing)
- [服務層級](https://docs.claude.com/zh-TW/api/service-tiers.md): 不同的服務層級讓您能夠根據應用程式的需求，在可用性、效能和可預測成本之間取得平衡。
- [支援的地區](https://docs.claude.com/zh-TW/api/supported-regions.md): 以下是我們目前可以支援存取的國家、地區和領土：
- [使用量和成本 API](https://docs.claude.com/zh-TW/api/usage-cost-api.md): 透過使用量和成本管理 API 程式化存取您組織的 API 使用量和成本資料。
- [版本](https://docs.claude.com/zh-TW/api/versioning.md): 在發出 API 請求時，您必須發送 `anthropic-version` 請求標頭。例如，`anthropic-version: 2023-06-01`。如果您使用我們的[客戶端 SDK](/zh-TW/api/client-sdks)，這將自動為您處理。
- [詞彙表](https://docs.claude.com/zh-TW/docs/about-claude/glossary.md): 這些概念並非 Anthropic 語言模型獨有，但我們在下方提供關鍵術語的簡要摘要。
- [模型淘汰](https://docs.claude.com/zh-TW/docs/about-claude/model-deprecations.md): 了解 Anthropic 模型的生命週期和淘汰時程表，包括遷移指南和最佳實踐。
- [選擇合適的模型](https://docs.claude.com/zh-TW/docs/about-claude/models/choosing-a-model.md): 為您的應用選擇最佳Claude模型需要平衡三個關鍵考量因素：能力、速度和成本。本指南幫助您根據特定需求做出明智的決定。
- [遷移至 Claude 4](https://docs.claude.com/zh-TW/docs/about-claude/models/migrating-to-claude-4.md): 本頁面提供從 Claude 3.7 模型遷移至 Claude 4 模型的指導。
- [模型概覽](https://docs.claude.com/zh-TW/docs/about-claude/models/overview.md): Claude 是由 Anthropic 開發的最先進大型語言模型系列。本指南介紹我們的模型並比較它們與舊版模型的性能。
- [定價](https://docs.claude.com/zh-TW/docs/about-claude/pricing.md): 了解 Anthropic 模型和功能的定價結構
- [內容審核](https://docs.claude.com/zh-TW/docs/about-claude/use-case-guides/content-moderation.md): 內容審核是維護數位應用程式中安全、尊重和高效環境的關鍵方面。在本指南中，我們將討論如何使用 Claude 來審核您數位應用程式中的內容。
- [客戶支援代理](https://docs.claude.com/zh-TW/docs/about-claude/use-case-guides/customer-support-chat.md): 本指南介紹如何利用 Claude 的進階對話能力來即時處理客戶查詢，提供 24/7 支援，減少等待時間，並以準確的回應和正面的互動來管理大量支援需求。
- [法律文件摘要](https://docs.claude.com/zh-TW/docs/about-claude/use-case-guides/legal-summarization.md): 本指南將介紹如何運用 Claude 的先進自然語言處理能力來高效摘要法律文件，提取關鍵資訊並加速法律研究。透過 Claude，您可以簡化合約審查、訴訟準備和監管工作，節省時間並確保法律流程的準確性。
- [常見使用案例指南](https://docs.claude.com/zh-TW/docs/about-claude/use-case-guides/overview.md)
- [工單路由](https://docs.claude.com/zh-TW/docs/about-claude/use-case-guides/ticket-routing.md): 本指南將介紹如何運用 Claude 的先進自然語言理解能力，根據客戶意圖、緊急程度、優先級、客戶檔案等因素，大規模分類客戶支援工單。
- [Google Sheets 附加元件](https://docs.claude.com/zh-TW/docs/agents-and-tools/claude-for-sheets.md): [Claude for Sheets 擴充功能](https://workspace.google.com/marketplace/app/claude%5Ffor%5Fsheets/909417792257) 將 Claude 整合到 Google Sheets 中，讓您可以直接在儲存格中執行與 Claude 的互動。
- [MCP 連接器](https://docs.claude.com/zh-TW/docs/agents-and-tools/mcp-connector.md): Claude 的模型上下文協議 (MCP) 連接器功能讓您能夠直接從 Messages API 連接到遠端 MCP 伺服器，無需單獨的 MCP 客戶端。
- [遠端 MCP 伺服器](https://docs.claude.com/zh-TW/docs/agents-and-tools/remote-mcp-servers.md): 遠端 MCP 伺服器
- [Bash 工具](https://docs.claude.com/zh-TW/docs/agents-and-tools/tool-use/bash-tool.md): Bash 工具讓 Claude 能夠在持續的 bash 會話中執行 shell 命令，允許系統操作、腳本執行和命令列自動化。
- [程式碼執行工具](https://docs.claude.com/zh-TW/docs/agents-and-tools/tool-use/code-execution-tool.md): Claude 可以分析資料、建立視覺化圖表、執行複雜計算、運行系統指令、建立和編輯檔案，並直接在 API 對話中處理上傳的檔案。
- [電腦使用工具](https://docs.claude.com/zh-TW/docs/agents-and-tools/tool-use/computer-use-tool.md): Claude 可以透過電腦使用工具與電腦環境互動，該工具提供螢幕截圖功能和滑鼠/鍵盤控制，實現自主桌面互動。
- [細粒度工具串流](https://docs.claude.com/zh-TW/docs/agents-and-tools/tool-use/fine-grained-tool-streaming.md): 細粒度工具串流
- [如何實作工具使用](https://docs.claude.com/zh-TW/docs/agents-and-tools/tool-use/implement-tool-use.md): 學習如何在 Claude 中實作工具使用功能，包括模型選擇、工具定義、控制輸出和處理錯誤。
- [使用 Claude 進行工具使用](https://docs.claude.com/zh-TW/docs/agents-and-tools/tool-use/overview.md)
- [文字編輯器工具](https://docs.claude.com/zh-TW/docs/agents-and-tools/tool-use/text-editor-tool.md): Claude 可以使用 Anthropic 定義的文字編輯器工具來查看和修改文字檔案，幫助您除錯、修復和改進您的程式碼或其他文字文件。
- [高效率代幣工具使用](https://docs.claude.com/zh-TW/docs/agents-and-tools/tool-use/token-efficient-tool-use.md): 了解如何使用 Claude Sonnet 3.7 的高效率代幣工具使用功能來節省代幣並減少延遲。
- [網頁擷取工具](https://docs.claude.com/zh-TW/docs/agents-and-tools/tool-use/web-fetch-tool.md): 網頁擷取工具允許 Claude 從指定的網頁和 PDF 文件中檢索完整內容。
- [網路搜尋工具](https://docs.claude.com/zh-TW/docs/agents-and-tools/tool-use/web-search-tool.md): 網路搜尋工具讓 Claude 能夠直接存取即時網路內容，允許它使用超越其知識截止點的最新資訊來回答問題。
- [批次處理](https://docs.claude.com/zh-TW/docs/build-with-claude/batch-processing.md)
- [引用](https://docs.claude.com/zh-TW/docs/build-with-claude/citations.md): Claude 能夠在回答有關文件的問題時提供詳細的引用，幫助您追蹤和驗證回應中的資訊來源。
- [上下文視窗](https://docs.claude.com/zh-TW/docs/build-with-claude/context-windows.md): 了解 Claude 的上下文視窗如何運作，包括擴展思考和工具使用的特殊行為。
- [嵌入向量](https://docs.claude.com/zh-TW/docs/build-with-claude/embeddings.md): 文本嵌入向量是文本的數值表示，能夠測量語義相似性。本指南介紹嵌入向量、其應用，以及如何使用嵌入模型進行搜索、推薦和異常檢測等任務。
- [使用延伸思考進行建構](https://docs.claude.com/zh-TW/docs/build-with-claude/extended-thinking.md)
- [Files API](https://docs.claude.com/zh-TW/docs/build-with-claude/files.md)
- [多語言支援](https://docs.claude.com/zh-TW/docs/build-with-claude/multilingual-support.md): Claude 在多種語言的任務中表現出色，相對於英語保持強大的跨語言表現。
- [功能概覽](https://docs.claude.com/zh-TW/docs/build-with-claude/overview.md): 探索 Claude 的進階功能和能力。
- [PDF 支援](https://docs.claude.com/zh-TW/docs/build-with-claude/pdf-support.md): 使用 Claude 處理 PDF。從您的文件中提取文字、分析圖表並理解視覺內容。
- [提示快取](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-caching.md)
- [保持清晰、直接和詳細](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/be-clear-and-direct.md)
- [讓 Claude 思考（思維鏈提示）以提升表現](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/chain-of-thought.md)
- [串接複雜提示以獲得更強大的表現](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/chain-prompts.md)
- [Claude 4 提示工程最佳實踐](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/claude-4-best-practices.md): Claude 4 模型的特定提示工程技術指南，幫助您在應用程式中獲得最佳結果。
- [延伸思考技巧](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/extended-thinking-tips.md)
- [長文本提示技巧](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/long-context-tips.md)
- [使用範例（多樣提示）來引導 Claude 的行為](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/multishot-prompting.md)
- [提示工程概述](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/overview.md)
- [預填 Claude 的回應以獲得更好的輸出控制](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/prefill-claudes-response.md)
- [自動生成初稿提示詞模板](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/prompt-generator.md): 使用我們的提示詞生成工具，為您的特定任務生成高品質的提示詞模板。
- [使用我們的提示詞改進器來優化您的提示詞](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/prompt-improver.md): 使用我們的提示詞改進器來優化您的提示詞
- [使用提示模板和變數](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/prompt-templates-and-variables.md)
- [使用系統提示詞賦予 Claude 角色](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/system-prompts.md)
- [使用 XML 標籤來結構化你的提示詞](https://docs.claude.com/zh-TW/docs/build-with-claude/prompt-engineering/use-xml-tags.md)
- [搜尋結果](https://docs.claude.com/zh-TW/docs/build-with-claude/search-results.md): 透過提供具有來源歸屬的搜尋結果，為 RAG 應用程式啟用自然引用功能
- [串流訊息](https://docs.claude.com/zh-TW/docs/build-with-claude/streaming.md): 學習如何使用伺服器發送事件 (SSE) 來串流 Claude 的回應，包括 SDK 使用方式和完整的 HTTP 串流回應範例。
- [Token 計數](https://docs.claude.com/zh-TW/docs/build-with-claude/token-counting.md): Token 計數讓您能夠在將訊息發送給 Claude 之前確定訊息中的 token 數量，幫助您對提示和使用做出明智的決定。
- [視覺](https://docs.claude.com/zh-TW/docs/build-with-claude/vision.md): Claude 3 和 4 系列模型具備新的視覺功能，讓 Claude 能夠理解和分析圖像，為多模態互動開啟令人興奮的可能性。
- [Amazon Bedrock 上的 Claude Code](https://docs.claude.com/zh-TW/docs/claude-code/amazon-bedrock.md): 了解如何透過 Amazon Bedrock 配置 Claude Code，包括設定、IAM 配置和故障排除。
- [分析](https://docs.claude.com/zh-TW/docs/claude-code/analytics.md): 查看您組織的 Claude Code 部署的詳細使用洞察和生產力指標。
- [CLI 參考](https://docs.claude.com/zh-TW/docs/claude-code/cli-reference.md): Claude Code 命令列介面的完整參考，包含命令和標誌。
- [常見工作流程](https://docs.claude.com/zh-TW/docs/claude-code/common-workflows.md): 了解 Claude Code 的常見工作流程。
- [有效管理成本](https://docs.claude.com/zh-TW/docs/claude-code/costs.md): 了解如何在使用 Claude Code 時追蹤和優化代幣使用量和成本。
- [資料使用](https://docs.claude.com/zh-TW/docs/claude-code/data-usage.md): 了解 Anthropic 對 Claude 的資料使用政策
- [開發容器](https://docs.claude.com/zh-TW/docs/claude-code/devcontainer.md): 了解 Claude Code 開發容器，適用於需要一致、安全環境的團隊。
- [Claude Code GitHub Actions](https://docs.claude.com/zh-TW/docs/claude-code/github-actions.md): 了解如何使用 Claude Code GitHub Actions 將 Claude Code 整合到您的開發工作流程中
- [Claude Code GitLab CI/CD](https://docs.claude.com/zh-TW/docs/claude-code/gitlab-ci-cd.md): 了解如何將 Claude Code 整合到您的 GitLab CI/CD 開發工作流程中
- [Google Vertex AI 上的 Claude Code](https://docs.claude.com/zh-TW/docs/claude-code/google-vertex-ai.md): 了解如何透過 Google Vertex AI 配置 Claude Code，包括設定、IAM 配置和故障排除。
- [Hooks 參考文件](https://docs.claude.com/zh-TW/docs/claude-code/hooks.md): 本頁面提供在 Claude Code 中實作 hooks 的參考文件。
- [開始使用 Claude Code hooks](https://docs.claude.com/zh-TW/docs/claude-code/hooks-guide.md): 學習如何透過註冊 shell 命令來自訂和擴展 Claude Code 的行為
- [身份與存取管理](https://docs.claude.com/zh-TW/docs/claude-code/iam.md): 了解如何為您的組織配置 Claude Code 的使用者驗證、授權和存取控制。
- [將 Claude Code 新增至您的 IDE](https://docs.claude.com/zh-TW/docs/claude-code/ide-integrations.md): 學習如何將 Claude Code 新增至您最愛的 IDE
- [互動模式](https://docs.claude.com/zh-TW/docs/claude-code/interactive-mode.md): Claude Code 會話中鍵盤快捷鍵、輸入模式和互動功能的完整參考。
- [法律與合規](https://docs.claude.com/zh-TW/docs/claude-code/legal-and-compliance.md): Claude Code 的法律協議、合規認證和安全資訊。
- [LLM 閘道配置](https://docs.claude.com/zh-TW/docs/claude-code/llm-gateway.md): 學習如何使用 LLM 閘道解決方案配置 Claude Code，包括 LiteLLM 設定、身份驗證方法，以及使用追蹤和預算管理等企業功能。
- [透過 MCP 將 Claude Code 連接到工具](https://docs.claude.com/zh-TW/docs/claude-code/mcp.md): 學習如何透過 Model Context Protocol 將 Claude Code 連接到您的工具。
- [管理 Claude 的記憶體](https://docs.claude.com/zh-TW/docs/claude-code/memory.md): 學習如何使用不同的記憶體位置和最佳實踐來管理 Claude Code 跨會話的記憶體。
- [模型配置](https://docs.claude.com/zh-TW/docs/claude-code/model-config.md): 了解 Claude Code 模型配置，包括像 `opusplan` 這樣的模型別名
- [監控](https://docs.claude.com/zh-TW/docs/claude-code/monitoring-usage.md): 了解如何為 Claude Code 啟用和配置 OpenTelemetry。
- [企業網路配置](https://docs.claude.com/zh-TW/docs/claude-code/network-config.md): 為企業環境配置 Claude Code，包括代理伺服器、自訂憑證授權機構 (CA) 和相互傳輸層安全性 (mTLS) 驗證。
- [輸出樣式](https://docs.claude.com/zh-TW/docs/claude-code/output-styles.md): 調整 Claude Code 以用於軟體工程以外的用途
- [Claude Code 概述](https://docs.claude.com/zh-TW/docs/claude-code/overview.md): 了解 Claude Code，Anthropic 的智能編程工具，它存在於您的終端中，幫助您比以往更快地將想法轉化為代碼。
- [快速入門](https://docs.claude.com/zh-TW/docs/claude-code/quickstart.md): 歡迎使用 Claude Code！
- [自訂工具](https://docs.claude.com/zh-TW/docs/claude-code/sdk/custom-tools.md): 建立並整合自訂工具以擴展 Claude Code SDK 功能
- [修改系統提示](https://docs.claude.com/zh-TW/docs/claude-code/sdk/modifying-system-prompts.md): 學習如何透過三種方法修改系統提示來自訂 Claude 的行為 - 輸出樣式、appendSystemPrompt 和 customSystemPrompt。
- [追蹤成本和使用量](https://docs.claude.com/zh-TW/docs/claude-code/sdk/sdk-cost-tracking.md): 了解並追蹤 Claude Code SDK 中用於計費的 token 使用量
- [無頭模式](https://docs.claude.com/zh-TW/docs/claude-code/sdk/sdk-headless.md): 以程式化方式執行 Claude Code，無需互動式 UI
- [SDK 中的 MCP](https://docs.claude.com/zh-TW/docs/claude-code/sdk/sdk-mcp.md): 使用 Model Context Protocol 伺服器透過自訂工具擴展 Claude Code
- [概述](https://docs.claude.com/zh-TW/docs/claude-code/sdk/sdk-overview.md): 使用 Claude Code SDK 建構自訂 AI 代理
- [處理權限](https://docs.claude.com/zh-TW/docs/claude-code/sdk/sdk-permissions.md): 在 Claude Code SDK 中控制工具使用和權限
- [Python SDK 參考](https://docs.claude.com/zh-TW/docs/claude-code/sdk/sdk-python.md): Claude Code Python SDK 的完整 API 參考，包括所有函數、類型和類別。
- [會話管理](https://docs.claude.com/zh-TW/docs/claude-code/sdk/sdk-sessions.md): 了解 Claude Code SDK 如何處理會話和會話恢復
- [SDK 中的斜線命令](https://docs.claude.com/zh-TW/docs/claude-code/sdk/sdk-slash-commands.md): 學習如何使用斜線命令透過 SDK 控制 Claude Code 會話
- [TypeScript SDK 參考](https://docs.claude.com/zh-TW/docs/claude-code/sdk/sdk-typescript.md): Claude Code TypeScript SDK 的完整 API 參考，包含所有函數、類型和介面。
- [串流輸入](https://docs.claude.com/zh-TW/docs/claude-code/sdk/streaming-vs-single-mode.md): 了解 Claude Code SDK 的兩種輸入模式以及何時使用每種模式
- [SDK 中的子代理](https://docs.claude.com/zh-TW/docs/claude-code/sdk/subagents.md): 在 Claude Code SDK 中使用子代理
- [待辦事項清單](https://docs.claude.com/zh-TW/docs/claude-code/sdk/todo-tracking.md): 使用 Claude Code SDK 追蹤和顯示待辦事項，實現有組織的任務管理
- [安全性](https://docs.claude.com/zh-TW/docs/claude-code/security.md): 了解 Claude Code 的安全防護措施和安全使用的最佳實踐。
- [Claude Code 設定](https://docs.claude.com/zh-TW/docs/claude-code/settings.md): 使用全域和專案層級設定以及環境變數來配置 Claude Code。
- [設定 Claude Code](https://docs.claude.com/zh-TW/docs/claude-code/setup.md): 在您的開發機器上安裝、驗證並開始使用 Claude Code。
- [斜線命令](https://docs.claude.com/zh-TW/docs/claude-code/slash-commands.md): 在互動式會話中使用斜線命令控制 Claude 的行為。
- [狀態列配置](https://docs.claude.com/zh-TW/docs/claude-code/statusline.md): 為 Claude Code 創建自定義狀態列以顯示上下文信息
- [子代理](https://docs.claude.com/zh-TW/docs/claude-code/sub-agents.md): 在 Claude Code 中創建和使用專門的 AI 子代理，用於特定任務的工作流程和改進的上下文管理。
- [優化您的終端機設定](https://docs.claude.com/zh-TW/docs/claude-code/terminal-config.md): Claude Code 在您的終端機正確配置時運作最佳。請遵循這些指南來優化您的體驗。
- [企業部署概述](https://docs.claude.com/zh-TW/docs/claude-code/third-party-integrations.md): 了解 Claude Code 如何與各種第三方服務和基礎設施整合，以滿足企業部署需求。
- [疑難排解](https://docs.claude.com/zh-TW/docs/claude-code/troubleshooting.md): 探索 Claude Code 安裝和使用常見問題的解決方案。
- [開始使用 Claude](https://docs.claude.com/zh-TW/docs/get-started.md): 進行您的第一次 Claude API 呼叫並建立一個簡單的網路搜尋助手
- [Claude 介紹](https://docs.claude.com/zh-TW/docs/intro.md): Claude 是由 Anthropic 建構的高效能、可信賴且智慧的 AI 平台。Claude 在涉及語言、推理、分析、編程等任務方面表現出色。
- [Anthropic 隱私權政策](https://docs.claude.com/zh-TW/docs/legal-center/privacy.md)
- [安全與合規](https://docs.claude.com/zh-TW/docs/legal-center/security-compliance.md)
- [使用 Claude 進行建構](https://docs.claude.com/zh-TW/docs/overview.md): Claude 是由 Anthropic 建構的一系列[高效能且智慧的 AI 模型](/zh-TW/docs/about-claude/models)。雖然 Claude 功能強大且可擴展，但它也是最值得信賴且可靠的 AI。它遵循關鍵協議，犯錯更少，並且抗拒越獄攻擊——讓[企業客戶](https://claude.com/customers)能夠大規模建構最安全的 AI 驅動應用程式。
- [定義你的成功標準](https://docs.claude.com/zh-TW/docs/test-and-evaluate/define-success.md)
- [建立強大的實證評估](https://docs.claude.com/zh-TW/docs/test-and-evaluate/develop-tests.md): 學習如何開發測試案例來衡量 LLM 性能並改進您的提示工程流程。
- [使用評估工具](https://docs.claude.com/zh-TW/docs/test-and-evaluate/eval-tool.md): [Claude Console](https://console.anthropic.com/dashboard) 具有**評估工具**功能，讓您可以在各種情境下測試您的提示。
- [null](https://docs.claude.com/zh-TW/docs/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals.md)
- [提高輸出一致性（JSON模式）](https://docs.claude.com/zh-TW/docs/test-and-evaluate/strengthen-guardrails/increase-consistency.md)
- [透過角色提示和預填來保持 Claude 的角色特性](https://docs.claude.com/zh-TW/docs/test-and-evaluate/strengthen-guardrails/keep-claude-in-character.md)
- [防範越獄和提示詞注入](https://docs.claude.com/zh-TW/docs/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks.md)
- [減少幻覺](https://docs.claude.com/zh-TW/docs/test-and-evaluate/strengthen-guardrails/reduce-hallucinations.md)
- [降低延遲](https://docs.claude.com/zh-TW/docs/test-and-evaluate/strengthen-guardrails/reduce-latency.md)
- [減少提示詞洩漏](https://docs.claude.com/zh-TW/docs/test-and-evaluate/strengthen-guardrails/reduce-prompt-leak.md)
- [null](https://docs.claude.com/zh-TW/home.md)
- [Claude 開發者平台](https://docs.claude.com/zh-TW/release-notes/api.md): 跟隨 Claude API 和 Claude Console 的更新。
- [Claude Apps](https://docs.claude.com/zh-TW/release-notes/claude-apps.md): 跟隨 Anthropic 的 Claude 應用程式的更新動態。
- [Claude Code](https://docs.claude.com/zh-TW/release-notes/claude-code.md): 隨時了解 Claude Code 的最新功能和改進
- [概覽](https://docs.claude.com/zh-TW/release-notes/overview.md): 跟隨 Anthropic 產品和服務的更新動態。
- [系統提示詞](https://docs.claude.com/zh-TW/release-notes/system-prompts.md): 查看 [Claude.ai](https://www.claude.ai) 和 Claude [iOS](http://anthropic.com/ios) 及 [Android](http://anthropic.com/android) 應用程式上核心系統提示詞的更新。
- [Claude 3.7 系統卡](https://docs.claude.com/zh-TW/resources/claude-3-7-system-card.md)
- [Claude 3 模型卡](https://docs.claude.com/zh-TW/resources/claude-3-model-card.md)
- [Claude 4 系統卡](https://docs.claude.com/zh-TW/resources/claude-4-system-card.md)
- [Anthropic 課程](https://docs.claude.com/zh-TW/resources/courses.md)
- [null](https://docs.claude.com/zh-TW/resources/overview.md)
- [適應性編輯器](https://docs.claude.com/zh-TW/resources/prompt-library/adaptive-editor.md): 根據使用者給定的指示重寫文本，例如使用不同的語調、受眾或風格。
- [機場代碼分析器](https://docs.claude.com/zh-TW/resources/prompt-library/airport-code-analyst.md): 從文本中查找並提取機場代碼。
- [外星人類學家](https://docs.claude.com/zh-TW/resources/prompt-library/alien-anthropologist.md): 從外星人類學家的角度分析人類文化和習俗。
- [頭韻煉金術師](https://docs.claude.com/zh-TW/resources/prompt-library/alliteration-alchemist.md): 為任何給定主題生成頭韻短語和句子。
- [Babel的廣播](https://docs.claude.com/zh-TW/resources/prompt-library/babels-broadcasts.md): 以世界上10種最常用語言創建引人注目的產品發布推文。
- [品牌建構師](https://docs.claude.com/zh-TW/resources/prompt-library/brand-builder.md): 為整體品牌識別制定設計簡報。
- [職業教練](https://docs.claude.com/zh-TW/resources/prompt-library/career-coach.md): 與AI職業教練進行角色扮演對話。
- [引用您的來源](https://docs.claude.com/zh-TW/resources/prompt-library/cite-your-sources.md): 獲得關於文件內容問題的答案，並提供支持回應的相關引用。
- [程式碼解釋器](https://docs.claude.com/zh-TW/resources/prompt-library/code-clarifier.md): 用簡單的語言簡化並解釋複雜的程式碼。
- [程式碼顧問](https://docs.claude.com/zh-TW/resources/prompt-library/code-consultant.md): 建議改進以優化 Python 程式碼效能。
- [企業預言家](https://docs.claude.com/zh-TW/resources/prompt-library/corporate-clairvoyant.md): 從冗長的企業報告中提取洞察、識別風險，並將關鍵資訊濃縮成單一備忘錄。
- [宇宙按鍵](https://docs.claude.com/zh-TW/resources/prompt-library/cosmic-keystrokes.md): 在單一 HTML 檔案中生成互動式速度打字遊戲，具有橫向捲動遊戲玩法和 Tailwind CSS 樣式。
- [CSV 轉換器](https://docs.claude.com/zh-TW/resources/prompt-library/csv-converter.md): 將各種格式（JSON、XML 等）的資料轉換為格式正確的 CSV 檔案。
- [烹飪創作者](https://docs.claude.com/zh-TW/resources/prompt-library/culinary-creator.md): 根據用戶可用的食材和飲食偏好建議食譜創意。
- [資料整理器](https://docs.claude.com/zh-TW/resources/prompt-library/data-organizer.md): 將非結構化文字轉換為客製化的 JSON 表格。
- [方向解碼器](https://docs.claude.com/zh-TW/resources/prompt-library/direction-decoder.md): 將自然語言轉換為逐步指導。
- [夢境解析師](https://docs.claude.com/zh-TW/resources/prompt-library/dream-interpreter.md): 為用戶的夢境提供象徵意義的解釋和洞察。
- [效率估算器](https://docs.claude.com/zh-TW/resources/prompt-library/efficiency-estimator.md): 計算函數和演算法的時間複雜度。
- [電子郵件提取器](https://docs.claude.com/zh-TW/resources/prompt-library/email-extractor.md): 從文件中提取電子郵件地址並整理成 JSON 格式的清單。
- [表情符號編碼器](https://docs.claude.com/zh-TW/resources/prompt-library/emoji-encoder.md): 將純文字轉換為有趣且富有表現力的表情符號訊息。
- [倫理困境導航器](https://docs.claude.com/zh-TW/resources/prompt-library/ethical-dilemma-navigator.md): 幫助使用者思考複雜的倫理困境並提供不同的觀點。
- [Excel 公式專家](https://docs.claude.com/zh-TW/resources/prompt-library/excel-formula-expert.md): 根據使用者描述的計算或資料操作建立 Excel 公式。
- [函數製造器](https://docs.claude.com/zh-TW/resources/prompt-library/function-fabricator.md): 根據詳細規格創建 Python 函數。
- [未來主義時尚顧問](https://docs.claude.com/zh-TW/resources/prompt-library/futuristic-fashion-advisor.md): 為用戶的特定偏好建議前衛的時尚趨勢和風格。
- [Git gud](https://docs.claude.com/zh-TW/resources/prompt-library/git-gud.md): 根據使用者描述的版本控制操作產生適當的 Git 指令。
- [Google apps 腳本編寫器](https://docs.claude.com/zh-TW/resources/prompt-library/google-apps-scripter.md): 根據用戶需求生成 Google Apps 腳本來完成任務。
- [評分專家](https://docs.claude.com/zh-TW/resources/prompt-library/grading-guru.md): 根據使用者定義的標準和準則比較和評估書面文本的品質。
- [語法精靈](https://docs.claude.com/zh-TW/resources/prompt-library/grammar-genie.md): 將語法錯誤的句子轉換為正確的英文。
- [Hal 幽默助手](https://docs.claude.com/zh-TW/resources/prompt-library/hal-the-humorous-helper.md): 與一個知識淺博且帶有諷刺性格的 AI 聊天。
- [成語解釋器](https://docs.claude.com/zh-TW/resources/prompt-library/idiom-illuminator.md): 解釋常見成語和諺語的含義和起源。
- [面試問題製作器](https://docs.claude.com/zh-TW/resources/prompt-library/interview-question-crafter.md): 為面試生成問題。
- [LaTeX 傳奇](https://docs.claude.com/zh-TW/resources/prompt-library/latex-legend.md): 撰寫 LaTeX 文件，生成數學方程式、表格等程式碼。
- [課程規劃師](https://docs.claude.com/zh-TW/resources/prompt-library/lesson-planner.md): 針對任何主題製作深入的課程計劃。
- [提示詞庫](https://docs.claude.com/zh-TW/resources/prompt-library/library.md)
- [主審核員](https://docs.claude.com/zh-TW/resources/prompt-library/master-moderator.md): 評估使用者輸入是否包含潛在有害或非法內容。
- [會議記錄員](https://docs.claude.com/zh-TW/resources/prompt-library/meeting-scribe.md): 將會議內容精煉為簡潔摘要，包括討論主題、關鍵要點和行動項目。
- [備忘錄大師](https://docs.claude.com/zh-TW/resources/prompt-library/memo-maestro.md): 根據關鍵要點撰寫全面的公司備忘錄。
- [正念導師](https://docs.claude.com/zh-TW/resources/prompt-library/mindfulness-mentor.md): 引導使用者進行正念練習和減壓技巧。
- [情緒色彩化器](https://docs.claude.com/zh-TW/resources/prompt-library/mood-colorizer.md): 將情緒的文字描述轉換為對應的十六進制色碼。
- [勵志繆思](https://docs.claude.com/zh-TW/resources/prompt-library/motivational-muse.md): 根據使用者輸入提供個人化的勵志訊息和肯定語句。
- [新詞創造者](https://docs.claude.com/zh-TW/resources/prompt-library/neologism-creator.md): 根據使用者提供的概念或想法發明新詞並提供其定義。
- [觀點思辨者](https://docs.claude.com/zh-TW/resources/prompt-library/perspectives-ponderer.md): 權衡用戶提供主題的利弊得失。
- [哲學沉思](https://docs.claude.com/zh-TW/resources/prompt-library/philosophical-musings.md): 參與深度哲學討論和思想實驗。
- [PII 淨化器](https://docs.claude.com/zh-TW/resources/prompt-library/pii-purifier.md): 自動檢測並從文本文件中移除個人身份識別資訊 (PII)。
- [多語言超能力](https://docs.claude.com/zh-TW/resources/prompt-library/polyglot-superpowers.md): 將任何語言的文字翻譯成任何語言。
- [混成詞詩人](https://docs.claude.com/zh-TW/resources/prompt-library/portmanteau-poet.md): 將兩個詞彙融合在一起，創造出一個新的、有意義的混成詞。
- [產品命名專家](https://docs.claude.com/zh-TW/resources/prompt-library/product-naming-pro.md): 根據描述和關鍵字創建吸引人的產品名稱。
- [散文潤飾師](https://docs.claude.com/zh-TW/resources/prompt-library/prose-polisher.md): 運用進階文案編輯技巧和建議來精煉和改善書面內容。
- [雙關語專家](https://docs.claude.com/zh-TW/resources/prompt-library/pun-dit.md): 根據任何給定主題生成巧妙的雙關語和文字遊戲。
- [Python 錯誤修復器](https://docs.claude.com/zh-TW/resources/prompt-library/python-bug-buster.md): 檢測並修復 Python 程式碼中的錯誤。
- [評論分類器](https://docs.claude.com/zh-TW/resources/prompt-library/review-classifier.md): 將回饋分類到預先指定的標籤和分類中。
- [猜猜這個謎語](https://docs.claude.com/zh-TW/resources/prompt-library/riddle-me-this.md): 生成謎語並引導用戶找到解答。
- [科幻情境模擬器](https://docs.claude.com/zh-TW/resources/prompt-library/sci-fi-scenario-simulator.md): 與使用者討論各種科幻情境以及相關的挑戰和考量。
- [二年級簡化器](https://docs.claude.com/zh-TW/resources/prompt-library/second-grade-simplifier.md): 讓複雜的文本變得容易讓年輕學習者理解。
- [比喻專家](https://docs.claude.com/zh-TW/resources/prompt-library/simile-savant.md): 從基本描述生成比喻。
- [蘇格拉底式智者](https://docs.claude.com/zh-TW/resources/prompt-library/socratic-sage.md): 就使用者提供的主題進行蘇格拉底式對話。
- [試算表魔法師](https://docs.claude.com/zh-TW/resources/prompt-library/spreadsheet-sorcerer.md): 生成包含各種類型資料的 CSV 試算表。
- [SQL 魔法師](https://docs.claude.com/zh-TW/resources/prompt-library/sql-sorcerer.md): 將日常語言轉換為 SQL 查詢。
- [故事創作夥伴](https://docs.claude.com/zh-TW/resources/prompt-library/storytelling-sidekick.md): 與用戶協作創造引人入勝的故事，提供情節轉折和角色發展。
- [時間旅行顧問](https://docs.claude.com/zh-TW/resources/prompt-library/time-travel-consultant.md): 幫助用戶探索假設性時間旅行情境及其影響。
- [繞口令](https://docs.claude.com/zh-TW/resources/prompt-library/tongue-twister.md): 創造具有挑戰性的繞口令。
- [問答題產生器](https://docs.claude.com/zh-TW/resources/prompt-library/trivia-generator.md): 產生各種主題的問答題，並在需要時提供提示。
- [推文語調檢測器](https://docs.claude.com/zh-TW/resources/prompt-library/tweet-tone-detector.md): 檢測推文背後的語調和情感。
- [VR 健身創新者](https://docs.claude.com/zh-TW/resources/prompt-library/vr-fitness-innovator.md): 為虛擬實境健身遊戲腦力激盪創意想法。
- [網站精靈](https://docs.claude.com/zh-TW/resources/prompt-library/website-wizard.md): 根據使用者規格建立單頁網站。
- [快速入門](https://docs.claude.com/zh-TW/resources/quickstarts.md): 快速開始使用 Claude 和 Claude API
- [系統狀態](https://docs.claude.com/zh-TW/resources/status.md)
