🤖 Claude Messages API

2 guides covering common problems, patterns, and production issues in Claude Messages API.

Building agents directly on Anthropic's Claude Messages API — the raw tool-use loop, multi-agent orchestration with subagents, and streaming, using the official `anthropic` SDK. (This covers hand-rolling the agent loop on the Messages API, distinct from Anthropic's separate higher-level Claude Agent SDK product.)

  • Tool use with a hand-managed agent loop (stop_reason, tool_use/tool_result blocks)
  • Multi-agent orchestration — orchestrator and subagent patterns
  • Streaming responses with real-time output
  • Model-agnostic across Claude Haiku, Sonnet, and Opus
  • Built on the official anthropic Python SDK
Visit official site →
Claude Messages API March 31, 2026 2 min

Building Your First Claude Agent: Tool Use, the Agent Loop, and Streaming

Most Anthropic SDK tutorials show a single tool call. They don't show what happens when Claude calls three tools in sequence, one tool fails, or the agent needs to decide whether to keep looping. That...

Read guide →
Claude Messages API March 31, 2026 2 min

Multi-Agent Systems with Claude: Orchestrators, Subagents, and When to Split

A single Claude agent with 20 tools and a 10,000-token task description works — until it doesn't. Long contexts degrade instruction-following. Too many tools dilutes selection quality. Tasks with clea...

Read guide →
Other Tools

Stay sharp as AI tools evolve

New guides drop regularly. Get them in your inbox — no noise, just signal.