Skip to content
Agentmart
Submit a tool

set-up-agents-md

Make `AGENTS.md` the single source of truth for a repository's agent instructions and symlink every agent's own instruction file to it.

Skill by MarvinJWendt

IndexedMIT★ 1
Write a review

Chalk Labs

Need help deploying set-up-agents-md?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/marvinjwendt-set-up-agents-md

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Make `AGENTS.md` the single source of truth for a repository's agent instructions and symlink every agent's own instruction file to it. Use when a repo has no agent instructions yet, when they live in `CLAUDE.md` as a real file, or when several instruction files have drifted apart.

Use cases: agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Nov 1, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/set-up-agents-md

What's inside

Skill name
set-up-agents-md
Description
Make `AGENTS.md` the single source of truth for a repository's agent instructions and symlink every agent's own instruction file to it. Use when a repo has no agent instructions yet, when they live in `CLAUDE.md` as a real file, or when several instruction files have drifted apart.
License
MIT
Instructions
39 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# set-up-agents-md

One instruction file per repo, `AGENTS.md` at the root. Each agent's own file is a symlink to it, so an instruction is written once and every agent reads the same text.

Claude is the default and usually the only agent to wire. Add another one (`GEMINI.md`) only when the user names it, using the same symlink. Codex and opencode read `AGENTS.md` directly and need nothing.

## Target layout

```
AGENTS.md                real file
CLAUDE.md -> AGENTS.md   symlink
```

Both sit in the same directory, so the symlink is a bare filename and works in every clone.

## Steps

1. Find what exists: `ls -la AGENTS.md CLAUDE.md GEMINI.md 2>&1`. A file that is already a symlink to `AGENTS.md` is done, leave it.
2. Bring the content together into `AGENTS.md`, by case:

| Found | Action |
| --- | --- |
| `AGENTS.md` only | Nothing to move. |
| `CLAUDE.md` only | `git mv CLAUDE.md AGENTS.md`, or `mv` when untracked. |
| Both | Read both. Merge every instruction from the other files into `AGENTS.md`, dropping only what it already says. Where the two contradict each other, ask the user which one holds. |
| Neither | Write one, see below. |

3. Replace each agent's file with the symlink: `rm CLAUDE.md && ln -s AGENTS.md CLAUDE.md`. Only after its content lives in `AGENTS.md`.
4. Verify the merge survived: `cat CLAUDE.md` reads back the instructions, and nothing that was in the old files is missing.

Repos carrying nested instruction files (`packages/foo/CLAUDE.md`) get the same treatment per directory, each symlink pointing at the `AGENTS.md` beside it.

Finish by naming which files merged and which are now symlinks.

## Writing the file from scratch

Read the repo first, enough to say what a new agent needs to know: build and test commands, the layout, the conventions the code follows but does not state. `writing-for-agents` owns how to write it. Keep it to what nearly every task in this repo needs, and let focused topics become skills through `set-up-skills`.

From MarvinJWendt/skills, MIT.

More from MarvinJWendt/skills

Signals

Stars
1
Forks
0
Open issues
0
Last commit (this folder)
Oct 10, 2026 (27 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from MarvinJWendt

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get set-up-agents-md running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying set-up-agents-md?

CallGet help
Cite this page

set-up-agents-md skill. Agent Mart. https://agentmart.dev/skills/marvinjwendt-set-up-agents-md (last updated 2026-10-24)