agents-md
Skill2 starsChanged 13 days agoMITView on GitHub
Create and maintain AGENTS.md files that provide AI coding agents with project context, commands, code style, and boundaries. Use when updating an existing AGENTS.md, creating a new AGENTS.md for a repository, or improving agent instructions for a codebase.
- Reads credentials
- Installs packages
36 other projects use a version of this.
Adopted by
Most recently taken up by
- caidanw/skillstheir own version, first seen Nov 8, 2026
- Integralist/agent-skillstheir own version, first seen Nov 6, 2026
- d-o-hub/github-template-ai-agentstheir own version, first seen Nov 3, 2026
- Rvey/agent-engineering-kittheir own version, first seen Nov 2, 2026
- Devolutions/IronRDPtheir own version, first seen Nov 1, 2026
- ytil/skillstheir own version, first seen Nov 1, 2026
Used with
Projects with it also use how, adr and 8 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
jaydubya818Skillhow
Use for runtime, ownership, layering, code-walkthrough, and placement questions.
Also in 3 projects that use this
asterkinSkilladr
Create Architecture Decision Records (ADRs) following project standards. Guides template usage, process, and best practices for documenting architecturally significant decisions.
Also in 3 projects that use this
jaydubya818Skillwhy
Use when design rationale must be recovered from source history and available records.
Also in 3 projects that use this
photon-groveSkillunslop
Detects and rewrites generic, overly polished, or AI-sounding prose while preserving meaning. Use when the user asks to "unslop", humanize, de-AI, tighten, or review whether writing sounds generated.
Also in 4 projects that use this
AWSSkillwait-what
Stop. That last message did not land: re-pitch it.
Also in 3 projects that use this
Harut8Skillarchitect
Also in 3 projects that use this
Other versions36 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 36
- agent-substrate/substrate★ 6.2ksame name
- Devolutions/IronRDP★ 3.3ksame name
- getsentry/skills★ 1.1ksame name
- mblode/agent-skills★ 189same name
- cumbucadev/cinemaempoa★ 145same name
- devbrother2024/skills★ 91same name
- Integralist/agent-skills★ 23same name
- arfar-x/agent-skills★ 14same name
- getsentry/webvitals.com★ 12same name
- caidanw/skills★ 11same name
26 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository11 other files it gives its agents.
- AGENTS.mdAGENTS.md
- chatgpt-share-markdownSkill
- ddgrSkill
- ghSkill
- jiraSkill
- jjSkill
- khalSkill
- notmuchSkill
- skill-creatorSkill
- taskwarriorSkill
- tmux-interactive-processesSkill
The file
We split this skill into the rules it gives an agent
7 rules. 0 are common, 7 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 32 edited versions we read keep these.
7
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 16
Create: When the project has non-obvious tooling, specific commands, or critical boundaries agents must know
- Distinctive: line 17
Update: When commands change, new critical constraints are added, or tooling requirements evolve
- Distinctive: line 18
Omit: If your repository has standard structure and common tooling (React, Python, etc.) that agents already understand
- Distinctive: line 70
Only document patterns that differ from standard conventions in your stack.
- Distinctive: line 150
If updating existing: Remove redundant sections, keep only high-value content
- Distinctive: line 151
Validate: Ensure every line provides information agents cannot infer
- Distinctive: line 231
Keep each file focused on what differs from defaults.
“Common” is measured: kept by at least 6 in 10 of the 32 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Agents Md
AGENTS.md is a README for AI agents: a dedicated place to provide context and instructions to help AI coding agents work effectively on a project.
Important: Keep It Minimal
Key principle: Include only what agents cannot infer from the codebase itself. Redundant documentation is harmful—agents already read files, search code, and understand common patterns.
When to Create/Update
- Create: When the project has non-obvious tooling, specific commands, or critical boundaries agents must know
- Update: When commands change, new critical constraints are added, or tooling requirements evolve
- Omit: If your repository has standard structure and common tooling (React, Python, etc.) that agents already understand
What to Include
Focus only on information agents cannot discover themselves:
1. Essential Commands (Required)
Specific commands with full flags. Include only if non-standard or critical.
## Commands
- Install: `pnpm install`
- Dev: `pnpm dev --port 3000`
- Test: `pnpm test --coverage`
- Build: `pnpm build`Skip this section if using standard commands (e.g., npm install, npm test).
2. Tooling Requirements (If Non-Standard)
Specify tools agents should use. Research shows agents DO follow specific tool instructions (e.g., uv usage increases 1.6x when mentioned).
## Tooling
- Use `uv` for Python dependency management
- Use `ruff` for linting (not flake8)
- Use `just` for task running (not Make)Only include if different from defaults agents would assume.
3. Boundaries (Critical)
What agents must NEVER do. This is the highest-value content.
## Boundaries
- Never commit secrets, API keys, or `.env` files
- Never modify `vendor/`, `node_modules/`, or lock files
- Never remove failing tests—fix or ask first
- Ask before changing database schemasAlways include this section. It prevents costly mistakes.
4. Project-Specific Patterns (If Unusual)
Only document patterns that differ from standard conventions in your stack.
## Patterns
- Use functional React components, not classes
- Place tests in `__tests__/` folders (not `.test.ts` files)
- Import order: React, third-party, `@/`, relativeSkip if following standard conventions for your framework/language.
What to Avoid
Based on research findings, do NOT include:
❌ Codebase Overviews
Research shows "context files do not provide effective overviews"—agents don't discover files faster with overviews present. They already search and explore code effectively.
Don't write:
## Overview
This is a React application with a component-based architecture...❌ Comprehensive Structure Documentation
Listing every directory is redundant. Agents can explore the filesystem.
Don't write:
## Structure
- `src/components/` - React components
- `src/hooks/` - Custom hooks
- `src/utils/` - Utility functions❌ Obvious Style Guidelines
Agents already know common patterns (camelCase, functional components, etc.).
Don't write:
## Code Style
- Use camelCase for functions
- Use PascalCase for components
- Prefer const over let❌ Verbose Explanations
Code examples are clearer than paragraphs. One snippet beats three paragraphs.
Prefer:
## Commands
- Test: `pytest -xvs tests/`Over:
## Testing
To run tests, use the pytest command. This will discover all tests in the tests directory and execute them with verbose output and fail-fast behavior...Workflow
- Check if needed: Does the repo have non-obvious tooling or critical constraints?
- If yes, create minimal file: Include only Commands (if non-standard), Tooling (if non-standard), Boundaries (always), and Patterns (if unusual)
- If updating existing: Remove redundant sections, keep only high-value content
- Validate: Ensure every line provides information agents cannot infer
Examples
Good: Minimal and Specific
# AGENTS.md
## Commands
- Install: `uv pip install -e ".[dev]"`
- Dev: `uvicorn app.main:app --reload --port 8000`
- Test: `pytest -xvs`
- Lint: `ruff check --fix .`
## Tooling
- Use `uv` for dependencies (not pip)
- Use `ruff` for linting and formatting
## Boundaries
- Never commit `.env` files
- Never modify `alembic/versions/` after merge
- Ask before schema changesBad: Redundant and Verbose
# AGENTS.md
## Overview
This is a FastAPI backend application with SQLAlchemy ORM.
## Tech Stack
- Python 3.12
- FastAPI 0.109
- SQLAlchemy 2.0
- PostgreSQL
## Structure
- `app/` - Application code
- `api/` - Route handlers
- `models/` - SQLAlchemy models
- `tests/` - Test files
## Code Style
- Use type hints on all functions
- Write docstrings for public functions
- Use async where possible
## Commands
- Install: `pip install -e ".[dev]"`
- Dev: `uvicorn app.main:app --reload`
- Test: `pytest`Why it's bad: Stack and structure are discoverable from files. Style guidelines are standard Python practices. Overview is unnecessary.
Monorepos
For monorepos, place minimal AGENTS.md files at subproject roots:
repo/
├── AGENTS.md (root - only if repo-wide constraints exist)
├── packages/
│ ├── api/
│ │ └── AGENTS.md (API-specific tooling/boundaries)
│ └── web/
│ └── AGENTS.md (web-specific tooling/boundaries)Keep each file focused on what differs from defaults.
References
- Best practices and anti-patterns: See references/best-practices.md
- Real-world examples: See references/examples.md
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

