LLM-driven systems that pursue a goal by interleaving reasoning, tool calls, and observations inside a loop — and that decide for themselves which step to take next.
The Anthropic Claude API — Python and TypeScript SDKs, streaming, tool use, batches, prompt caching, files, vision, and agentic loops.
Define tools, handle tool calls, run agentic loops, use parallel tools, and manage errors with the Claude API.
Complete TypeScript SDK reference for the Anthropic Claude API — install, messages.create, streaming, tool use, vision, types, and Node/Deno/Bun integration.
Techniques for reliable structured generation — JSON mode, schema-constrained decoding, function/tool calls as output, and validator pairing with Pydantic or Zod.
navigation
actions
cheat sheet pages