This week: 18% of 499 scanned repos had AGENTS.md.See Radar #4 →

What it is

AGENTS.md is a standard file that many AI agents read to learn how to work in your repository — build commands, test commands, conventions, and constraints.

Why use it

A single AGENTS.md works across multiple tools that support the convention, so you maintain one set of instructions instead of per-tool files. You can use it to document hygiene rules for agents.

How to set it up

Create AGENTS.md in your project root and describe how agents should build, test, and behave, including any sensitive paths to avoid.

Example AGENTS.md
# AGENTS.md

## Conventions
Use pnpm. Run `pnpm test` before finishing.

## Off-limits
Never read .env or files under secrets/.

Official AI agents (open standard) docs →

What to exclude

Common sensitive categories worth adding to AGENTS.md:

FAQ

Which tools read AGENTS.md?
A growing set of AI coding agents support the convention. Check your tool’s docs, and combine AGENTS.md with a dedicated ignore file for enforcement.

Check your own project

158 of 499 repos this week exposed sensitive paths to AI tools. Check yours in seconds:

Run the same local check with Offsend CLI:

brew install --cask offsend/tap/offsend-cli
offsend show
offsend prepare --dry-run

Everything runs locally. No file contents are uploaded.