agentleFS
Sign inStart free
NeplichNeplich/dev-agent-skills/AGENTS.md

dev-agent-skills

AGENTS.md8 starsChanged 2 months agoMITView on GitHub

This repository publishes an on-demand professional knowledge library: six plugins and thirty-seven directly usable Skills. Role names organize expertise. The current assistant owns the user's task and combines relevant capabilities through completion. Use the user's request, issues, code, tests and existing documents to establish the expected outcome. Select references, plans, formal documents and sub-agents when they materially help the task. Continue within existing authorization; ask when a material decision is missing or an action needs additional permission. Repository Skill content…

  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 31,439 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repository3 other files it gives its agents.

The file

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

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use the manual release procedure.

    line 27
  2. Very specific:

    Prefer uv for Python and Chrome-based tools for browser interaction.

    line 31

≈ “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

Repository Instructions

This repository publishes an on-demand professional knowledge library: six plugins and thirty-seven directly usable Skills. Role names organize expertise. The current assistant owns the user's task and combines relevant capabilities through completion.

Working Approach

Use the user's request, issues, code, tests and existing documents to establish the expected outcome. Select references, plans, formal documents and sub-agents when they materially help the task. Continue within existing authorization; ask when a material decision is missing or an action needs additional permission.

Repository Skill content is the product being maintained. Maintain this repository using the active assistant's configuration and the user's instructions. Preserve unrelated user changes, credentials and accurate verification evidence. Keep final artifacts and remove task-created temporary files and caches at completion.

Source Layout

  • Plugins live in agents/{role}/, with bilingual READMEs and skills/.
  • Each SKILL.md describes a directly usable capability; _internal/ and references/ provide optional detail.
  • .claude-plugin/marketplace.json, plugin manifests and skills-lock.json describe installation and distribution.
  • Shared references are maintained in agents/product_manager/skills/idea-to-spec/_internal/_shared/. Generate their packaged copies with scripts/generate_shared_contracts.py.
  • AGENTS.md is the repository guidance source; CLAUDE.md is its relative symlink.

See architecture, document guidance and Skill maintenance for the relevant details.

Git and Release

Make maintenance changes on a branch and deliver them through a PR. Use <type>(<scope>): <中文描述> for commit and PR titles, Chinese bodies, and the author's own attribution. Preserve existing commits after a PR is created; append and push new commits normally. Amend, rebase and force-push require an explicit user request.

Merge after explicit maintainer authorization, using squash by default. After a verified merge, switch to the default branch, pull with --ff-only, and delete the exact merged local branch. Remote branch deletion requires explicit authorization.

Use the manual release procedure. Publication and deployment follow the user's explicit authorization. GitHub operations prefer Connector, then authenticated gh / GraphQL, then Chrome; verify bulk writes with gh.

Verification

Prefer uv for Python and Chrome-based tools for browser interaction. Read local time for time-sensitive work. Run relevant behavior tests together with:

uv run scripts/generate_shared_contracts.py --check
uv run scripts/check_repository_contract.py
uv run scripts/check_doc_contract.py
git diff --check

Report what changed, the verification results and any remaining limitations.

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