agentleFS
Sign inStart free
calven-aicalven-ai/marketing-as-code/.agents/skills/propose/SKILL.md

propose

Skill10 starsChanged 39 days agoMITView on GitHub

Turn what changed into a proposal (a pull request) and hand back the link. Use when asked to "propose", "send this for review", "ship it", or when work is done and should land.

  • Deletes or force-pushes
  • Commits and pushes

10 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 9 edited versions we read keep these.

  • 1

    Distinctive

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

    Scope: if the files include something the person did not work on in this session, say so and ask whether it belongs in this proposal.

    line 24
  2. Very specific:

    Never run git push --force, gh pr merge, gh pr review, or push to main.

    line 46
  3. Distinctive:

    Never include credentials or a customer's personal data.

    line 52

“Common” is measured: kept by at least 6 in 10 of the 9 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.

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

Propose

The ship-it word. Saying it is the consent: the proposal opens in the same turn, with no second question. The script does the deterministic part (check, fix, commit, push, open); this skill does the judgment (scope, the title, the description, the fixes the script cannot make).

Procedure

  1. Look first: python3 scripts/propose.py --check --json. It lists the files, the findings and the kind (bookkeeping or needs review).
  2. Scope: if the files include something the person did not work on in this session, say so and ask whether it belongs in this proposal. --paths leaves things out. Otherwise proceed without asking.
  3. Fix what the script cannot: for each remaining finding, fix it in the file (a frontmatter value, a broken link, a misplaced file moved with git mv), then run --check again. A finding about something shaped like a key: stop, say so, and point at docs/secrets.md.
  4. Write the title and the description. Title in sentence case, under 80 characters, "Subject: what changed"; it becomes the commit title on the approved copy. Description: two short paragraphs, what changed and why, in plain words, naming the brief or decision behind it when there is one. Save the description to a temporary file.
  5. Run it: python3 scripts/propose.py --title "..." --body-file <file>.
  6. Relay: the link, the outcome sentence (merges itself, you are the reviewer, or waiting for someone), and that the check takes about a minute. If the script printed a "Create pull request" link instead, say to open it and click the button.
  7. If it refused: relay the reason in the script's words and the one action it names. Nothing was sent.

Rules

  • Never run git push --force, gh pr merge, gh pr review, or push to main. The script does not either.
  • Content goes through /review before it is proposed as in-review or published; suggest it when the proposal holds a draft. A draft that flips to published in this proposal must carry published (the date) and published_url; confirm both before running.
  • Never include credentials or a customer's personal data.

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