Heads up This site is currently under heavy development.
← all datasets
◆ Agent instructions

AGENTS.md

MIT

The open AGENTS.md format for guiding coding agents, with a sample file — the cross-vendor counterpart to CLAUDE.md and .cursorrules.

Summary

AGENTS.md is a simple, open format for guiding coding agents. Think of AGENTS.md as a README for agents: a dedicated, predictable place to provide context and instructions to help AI coding agents work on your project.
from the project README
Formats
agent-md
License
MIT
Added
2026-09-10

Get it

From the project’s own instructions where it documents any; otherwise a plain clone.

  1. shell
    $ git clone --depth 1 https://github.com/agentsmd/agents.md

    Shallow clone: history isn’t needed for testing.

Contents

62 files 2.1 MB repository

  • .svg 27
  • .tsx 19
  • .png 3
  • .json 2
  • .md 2
  • (no ext) 2
  • .ts 2
  • .css 1

Use it with

Commands are curated, not yet run by us.

Claude Code — Install one subagent definition for every project on this machine.

agent.md
$ cp agent.md ~/.claude/agents/

Claude Code — Confirm the subagent loaded.

agent.md
$ claude -p 'List the subagents you can delegate to and what each one is for.'

OpenAI Codex CLI — Have Codex read an instruction file non-interactively.

agent.md
$ codex exec 'Read agent.md and summarise the instructions it gives an agent.'
my-toolchain — 0 tools
paste an install list to detect your tools

A brew list, a Brewfile, requirements.txt, a Dockerfile — or just the product names, free-form. Nothing leaves your browser.

    browse all tools →

    Send us feedback