agentleFS
Sign inStart free
probabl-aiprobabl-ai/skore-skills/skills/persist-ml-git/SKILL.md

persist-ml-git

Skill144 starsChanged 17 days agoBSD-3-ClauseView on GitHub

Commit the current loop stage when the end-turn hook says invoke. Trigger after `python -m skore_skills git end-turn --stage <stage>` returns action invoke, or when the user asks to commit this turn's work.

  • Reads credentials
  • Commits and pushes

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 1 is distinctive, and 7 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 7

    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:

    If action is skip, stop. Do not nag.

    line 15
  2. Very specific:

    If reason is resolve-dotfiles and ambiguous_dotfiles is non-empty, ask once which hidden paths to keep.

    line 17
  3. Very specific:

    If review_paths is non-empty, ask once which of those paths to keep tracked.

    line 31
  4. Very specific:

    Run git status. Stage only non-secret paths that are not still listed in review_paths: git add -- <paths>.

    line 51
  5. Very specific:

    Commit with a one-line subject from this turn's files and intent: git commit -m "…".

    line 54
  6. Distinctive:

    Do not git push, amend, rebase, or git config.

    line 60
  7. Very specific:

    Do not invent a commit when the hook was skip (off, unanswered, no_repo, clean).

    line 61
  8. Very specific:

    Do not call python -m skore_skills git end-turn --stage <stage> to perform the commit.

    line 63

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill 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
SKILL.md

Persist ML Git

The hook already ran this turn. Follow its JSON. Run git yourself.

  1. If action is skip, stop. Do not nag. Load triage-ml-task if installed, else stop.
  2. If reason is resolve-dotfiles and ambiguous_dotfiles is non-empty, ask once which hidden paths to keep. Include an explicit Keep none option. An empty multi-select is not an answer. Then python -m skore_skills git ignore-merge --decide plus --keep <path> for each chosen path (no --keep if they keep none, including when they pick Keep none). Do not re-ask names gone from the next JSON. Never keep .env or .skore. If ambiguous_dotfiles is empty or reason is persist, do not ask about hidden paths. Exit code 2 from git ignore-merge is the structured dotfile decision required outcome, not a generic command failure: read its JSON, ask the decision, and rerun with --decide.
  3. If review_paths is non-empty, ask once which of those paths to keep tracked. Include an explicit Keep none option. An empty multi-select is not an answer. List each JSON path and kind, and say that the rest are ignored. Then python -m skore_skills git review-decide with --keep <path> for each chosen path and --ignore <path> for every other path in the list (no --keep if they keep none, including when they pick Keep none). Pass a folder path through with its trailing slash so the ignore line is that folder. Copy paths and kinds from the JSON; do not invent them. Do not re-ask paths gone from the next JSON. Never --keep .env or .skore. If review_paths is empty, do not ask about file size, artifacts, or datasets. Do this when reason is resolve-review, and also after step 2 when reason is resolve-dotfiles and that payload lists review_paths. Exit code 2 from git review or git review-decide is the structured review decision required outcome, not a generic command failure: read its JSON, ask the decision, and rerun review-decide.
  4. Run git status. Stage only non-secret paths that are not still listed in review_paths: git add -- <paths>. Never .env or .skore.
  5. Commit with a one-line subject from this turn's files and intent: git commit -m "…". Do not use a canned stage slogan.
  6. Load triage-ml-task if installed, else stop.

Stop conditions

  • Do not git push, amend, rebase, or git config.
  • Do not invent a commit when the hook was skip (off, unanswered, no_repo, clean).
  • Do not call python -m skore_skills git end-turn --stage <stage> to perform the commit; that command only prints facts.

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