Claude API
The Claude API is the HTTP surface + Python/TypeScript SDKs you use when you want to build your own apps. Tool use, structured outputs, streaming, vision, prompt caching, the message batches API, the files API, the agent skills surface. Pricing is per-token; the Sonnet 4.6 tier hits the right cost/quality balance for most work today (verify before committing — model lineup changes faster than Claw can keep up). All entries below are sourced from docs.anthropic.com/api.
- API.1 Claude API, plainly sourced
Five-minute orientation to the Claude API — what shape it takes, what it can do (tool use, structured outputs, vision, streaming, prompt caching), what it costs, and where it fits vs Claude Code or Claude.ai.
- API.2 Getting started with the Claude API sourced
From zero to a working API call in 5 minutes. Get a key, install the SDK, make your first request in Python or TypeScript, handle the response, and set a billing cap.
- API.3 Claude models — Sonnet, Opus, Haiku sourced
What each Claude tier is good for, what they cost in relative terms, how to pick between them, and the always-changing context window + vision support matrix. Pricing and model names drift fast — verify before committing.
- API.4 Common patterns — tool use, structured outputs, caching sourced
Four patterns that come up over and over building on the Claude API — the tool-use loop, structured JSON outputs, prompt caching for long system prompts, and message batches for bulk async work. Code in Python + TypeScript.