agentleFS
Sign inStart free
robcsaszarrobcsaszar/prose/AGENTS.md

prose

AGENTS.md0 starsChanged 2 months agoMITView on GitHub

This repo publishes the prose skill for others to install. There is no build; the deliverable is the contents of skills/prose/. The scanner has a test suite. After any change to scripts/slop-scan.py, run it from skills/prose/: tests/fixtures/clean-blog.md and the lower half of tests/fixtures/noun-stack.md are ordinary prose that must stay silent. If a rule change makes those cases fail, the rule is wrong — do not update the golden to match. Regenerate goldens with --update only when you intended the behaviour…

  • 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 31,774 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 repositoryOne other file it gives its agents.

The file

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

4 rules. About 0 are common, 3 are distinctive, and 1 is 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.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Don't let the skill's content drift toward being specific to any one project.

    line 19
  2. Very specific:

    Add a ## [X.Y.Z] - YYYY-MM-DD block at the top of CHANGELOG.md, and its [X.Y.Z]: … link reference at the bottom.

    line 35
  3. Distinctive:

    Never push a tag or create a release outside the workflow.

    line 47
  4. Distinctive:

    Never work around a failed run by hand-writing notes or skipping a check.

    line 48

≈ “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

Mission

This repo publishes the prose skill for others to install. There is no build; the deliverable is the contents of skills/prose/.

The scanner has a test suite. After any change to scripts/slop-scan.py, run it from skills/prose/:

python3 tests/run-tests.py

tests/fixtures/clean-blog.md and the lower half of tests/fixtures/noun-stack.md are ordinary prose that must stay silent. If a rule change makes those cases fail, the rule is wrong — do not update the golden to match. Regenerate goldens with --update only when you intended the behaviour change.

Judgment boundaries

ASK:

  • Don't let the skill's content drift toward being specific to any one project; it needs to read as generic writing guidance. Ask before changing the license or copyright holder.

Releasing

Releases are cut by the Release workflow (.github/workflows/release.yml), never by hand. It is a manual workflow_dispatch with one input, tag, and it releases the commit at the tip of the branch it is run on. Run it on main.

The workflow, in order:

  1. Rejects a tag that is not vX.Y.Z, or that already exists.
  2. Fails unless version in .claude-plugin/plugin.json and plugins[0].version in .claude-plugin/marketplace.json both equal X.Y.Z.
  3. Takes the ## [X.Y.Z] block from CHANGELOG.md as the release notes, and fails if there is none.
  4. Creates the tag at the checked-out commit and publishes the GitHub release with those notes.

Nothing is created until every check passes, so a failed run leaves nothing to clean up.

To prepare a release, in one PR:

  • Set the same new version in .claude-plugin/plugin.json and .claude-plugin/marketplace.json.
  • Add a ## [X.Y.Z] - YYYY-MM-DD block at the top of CHANGELOG.md, and its [X.Y.Z]: … link reference at the bottom.
  • Merge to main.

Then run the workflow on main with tag=vX.Y.Z, from the Actions tab (Release → Run workflow) or from a shell:

gh workflow run release.yml --ref main -f tag=vX.Y.Z

Afterwards, confirm the release exists and its notes match the changelog block.

NEVER:

  • Never push a tag or create a release outside the workflow. A hand-made tag makes the workflow refuse that version, and a hand-written release skips the changelog and version checks.
  • Never work around a failed run by hand-writing notes or skipping a check. Fix the changelog or the manifests, merge, and re-run.

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