# Claude API The RESTful API at `api.anthropic.com`. It's the programmatic way to use [[Claude]]: send a request, get a response. The API is the core of the [[Claude Platform]]; every SDK, every integration, [[Claude Code]], and [[Claude Managed Agents]] all sit on top of it. ## What's in it Generally available: - **Messages API** — the main one; send messages, get conversational responses (`POST /v1/messages`) - **Message Batches** — process large volumes asynchronously at 50% off - **Token Counting** — count tokens before sending, so cost and rate limits don't surprise you - **Models** — list the available Claude models and their details Beta: - **Files** — upload a file once, reuse it across many calls - **Skills** — create and manage custom [[AI Agent Skills|agent skills]] - **Agents / Sessions / Environments** — the [[Claude Managed Agents]] endpoints ## Authentication Every request carries an `x-api-key` header (your key from the [[Claude Console]]) plus an `anthropic-version`. The SDKs add these for you. For machine-to-machine auth without static keys, there's Workload Identity Federation, which swaps a short-lived token for a bearer credential. ## SDKs over raw HTTP You can hit the endpoints with any HTTP client. But the official SDKs (Python, TypeScript, Java, Go, C#, Ruby, PHP) handle headers, retries, streaming, and typing for you. Use them unless you have a reason not to. ## Direct vs cloud The direct API gives you the newest models first and a direct line to Anthropic for billing and support. The same models also run on AWS, Vertex AI, and Azure if your infrastructure already lives there, though feature availability and request size limits vary by platform. ## References - API overview: https://platform.claude.com/docs/en/api/overview - Get API keys: https://platform.claude.com/settings/keys - Client SDKs: https://platform.claude.com/docs/en/api/client-sdks ## Related - [[Claude Platform]] - [[Claude Console]] - [[Claude]] - [[Claude Code]] - [[Claude Managed Agents]] - [[Anthropic SDK]] - [[Anthropic CLI (ant)]] - [[AI Agent Skills]] - [[Large Language Models (LLMs)]]