Pi Coding Agent: Four Built-In Tools (read, bash, edit, write)

Melvin Vivas · X post · 2026-08-10 · Open on X

Topics: AI Agents, Tool Use & MCP, AI Dev Tools & Productivity · Level: intermediate

Summary

The quoted post explains that the Pi coding agent ships with only four tools: read, bash, edit and write. It argues these are mostly all a coding agent needs, because models already know how to use bash well. Pi's SDK lets you add these same primitives to your own app.

Key points

• Pi's four built-in tools are read, bash, edit and write.
• A small set of basic tools is enough for an effective coding agent.
• Models are good at using bash and can do more with these basics than you might expect.
• Pi's SDK lets you add these tool primitives to your own application.

Resources mentioned

• Pi (pi.dev) · tool · pi.dev · free
  Minimal coding agent with four core tools, plus an SDK for adding those tools to your own apps.
  Also in: Re-authenticate the Pi Agent with the New ChatGPT Sign-In (Melvin Vivas on X · notes), Coworker: Open-Source Work Agent That Runs on Small Local Models (Melvin Vivas on X · notes), Running Coworker on Free OpenRouter Models and Its Agent Stack (Melvin Vivas on X · notes)

Try this

• Try building a coding agent that uses only read, bash, edit and write tools, for example with Pi's SDK.
• Add a minimal coding agent with Pi's four primitive tools to your own app.