agentleFS
Sign inStart free
itsryanarmaniitsryanarmani/skills/AGENTS.md

skills

AGENTS.md1 starsChanged 4 days agoMITView on GitHub

Read .agents/brand.md before changing the repository identity, public copy, skill names, or claims. Read .agents/design.md before changing the README, documentation structure, or the presentation of a skill. When the task touches directory branding, /design, or public kit components: Install upstream once per machine: design-taste ./scripts/install.sh; better-design npx skills add marvkr/better-design. This repository is public. Keep private repository names, project domains, private links, and identifying pull request or run numbers out of skills. Use generic examples, such as <your-package> or "a…

  • Reads credentials

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this AGENTS.md into the rules it gives an agent

8 rules. About 0 are common, 3 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Very specific:

    Read .agents/brand.md before changing the repository identity, public copy, skill names, or claims.

    line 5
  2. Very specific:

    Read .agents/design.md before changing the README, documentation structure, or the presentation of a skill.

    line 8
  3. Very specific:

    When the task touches directory branding, /design, or public kit components:

    line 13
  4. Very specific:

    Use generic examples, such as <your-package> or "a multi-site monorepo".

    line 25
  5. Distinctive:

    Put named fleet reports in the private operations repository.

    line 27
  6. Distinctive:

    Write "Fix the drop-off", not "Fixed the drop-off".

    line 47
  7. Distinctive:

    Run the project's test/lint command before opening a PR (see below)

    line 78
  8. Very specific:

    Follow the fleet PR standard in this repo's AGENTS.md (<!-- pr-standards:start --> block).

    line 88

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

AGENTS.md

Brand and design context

Read .agents/brand.md before changing the repository identity, public copy, skill names, or claims.

Read .agents/design.md before changing the README, documentation structure, or the presentation of a skill.

Directory & product UI (multi-site directory monorepos)

When the task touches directory branding, /design, or public kit components:

  1. directory-fleet-ui — presets, variants, fleet brand pass
  2. design-taste — anti-slop; install upstream pack; use corporate-trust + civic for reference sites
  3. better-design — MCP review; map catalog systems to presets, never vendor shadcn into apps/web

Install upstream once per machine: design-taste ./scripts/install.sh; better-design npx skills add marvkr/better-design.

Public naming boundary

This repository is public. Keep private repository names, project domains, private links, and identifying pull request or run numbers out of skills. Use generic examples, such as <your-package> or "a multi-site monorepo". Public dependencies keep their published package identifiers so instructions work. Put named fleet reports in the private operations repository.

<!-- pr-standards:start -->

Pull requests

One issue. One PR. One concern. Under 500 counted lines.

Open the issue first. No issue, no branch. The issue number ties the branch, the title, the body and the merged commit to one agreed piece of work.

branch:  ski-<issue>-<slug>          ski-142-fix-onboarding-drop-off
title:   [SKI-<issue>] <Subject>   [SKI-142] Fix onboarding drop-off
body:    Closes #142
         ## What / ## Why / ## How I verified
         Assisted-by: <agent>:<model>

Subject line: imperative mood, 10-50 characters, no trailing period, no emoji. Write "Fix the drop-off", not "Fixed the drop-off".

Hard caps, failed by the pr-standards CI check: 500 counted lines, 40 counted files, exactly one Closes #. Lockfiles, build output, snapshots, generated code and migrations are not counted. There is no label that clears the cap and no one to ask for one. Split the change.

Settings for this repo are in .github/pr-standards.json. The standard is at https://github.com/itsryanarmani/scripts/blob/main/pr-standards.md

<!-- pr-standards:end --><!-- cursor-cloud:start -->

Cloud agents (Cursor)

This repo runs on Cursor Cloud Agents. Local .env.local does not sync — mirror keys in Dashboard → Cloud Agents → Secrets.

Secret typeUse for
Runtime SecretAPI keys, passwords (hidden from chat/commits)
Environment VariableNon-sensitive config (URLs, flags)
Build SecretPrivate npm/docker registries during install only

Install & test

Install command lives in .cursor/environment.json. After dashboard setup:

  1. Environments → link this repo → wait for Build = Success
  2. Secrets → copy every key from your local .env.local / .env.example
  3. Run the project's test/lint command before opening a PR (see below)

Verify before PR

echo 'No test command detected'

Pull requests

Follow the fleet PR standard in this repo's AGENTS.md (<!-- pr-standards:start --> block). Cloud agents need push access via Git integration and a successful environment Build.

Setup guide: https://github.com/itsryanarmani/scripts/blob/main/cursor-cloud-rollout.md

<!-- cursor-cloud:end -->

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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.

Check your setup, free