Build with a Coding Agent

Build your Pipecat agent with Claude Code or Codex, fed live, accurate context.

Pipecat works with AI coding tools like Claude Code and Codex to write your agent code. Make your project agent-ready, give the agent live context instead of stale training data, then let it build.

1. Make your project agent-ready

   Install the CLI and run pipecat init in your project:

   uv tool install "pipecat-ai[cli]"
   pipecat init
   

   This writes a Pipecat coding-agent guide (AGENTS.md + CLAUDE.md) and developer guidance (GETTING_STARTED.md). Your agent picks up AGENTS.md automatically — it follows Pipecat conventions and scaffolds the app for you with pipecat create.

   pipecat init reference

   What init writes and how the agent flow works

2. Add the Pipecat Context Hub

   The Context Hub indexes Pipecat docs, examples, and API source into a local database, so your agent queries live context instead of stale training data. Build the index, then add it as an MCP server:

   uvx pipecat-ai-context-hub@latest refresh   # build local index (~2 min first run)
   

   Claude Code

   claude mcp add pipecat-context-hub -- uvx pipecat-ai-context-hub serve
   

   Codex

   codex mcp add pipecat-context-hub -- uvx pipecat-ai-context-hub serve
   

   Pipecat Context Hub reference

   Full setup, the tool list, CLI lookups, and MCP config for Cursor and VS Code

3. Start a coding session

   Open Claude Code or Codex in your project and prompt it to build. The agent follows AGENTS.md, queries the Context Hub for accurate APIs, and scaffolds and iterates on your bot.

Build Your Next Bot

The full agent-driven workflow, start to finish

Pipecat Context Hub

Everything the Context Hub indexes and exposes to your agent

Prefer to feed context by hand? Pipecat docs support the llms.txt standard: llms.txt is a structured index of all pages, and llms-full.txt is the full documentation in a single file — useful for tools that ingest docs in bulk.