agentleFS
sabirmgdsabirmgd/terminal-first-goal-loop/skills/pr-manager/SKILL.md

pr-manager

Skill2 starsChanged 2 months agoMITView on GitHub

Manage the pull-request or merge-request lifecycle from exact repo and branch resolution through draft creation, updates, monitoring, and ready-for-review transitions, without merging or posting without authority.

2 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

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

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

    Use this when code changes need a pull request or merge request, or when an existing review should be monitored from the terminal.

    line 8
  2. Distinctive:

    Before any action, resolve:

    line 26
  3. Distinctive:

    Do not guess the repo from the current shell if the task can name it explicitly.

    line 37
  4. Distinctive:

    create or update the PR or MR

    line 66
  5. Distinctive:

    read the saved result back

    line 67
  6. Distinctive:

    Never merge from this skill without explicit authority.

    line 87
  7. Distinctive:

    Never post comments or review requests without explicit authority.

    line 88
  8. Very specific:

    Keep the repo/head/base exact.

    line 89
  9. Distinctive:

    Do not call a branch ready just because local tests passed if required remote checks are still pending.

    line 90

≈ “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

PR Manager

Use this when code changes need a pull request or merge request, or when an existing review should be monitored from the terminal.

Default mode

Read-only by default.

Safe default actions:

  • inspect an existing PR or MR
  • resolve exact repo, head, and base
  • list checks, comments, or merge state
  • draft a PR body
  • prepare an update or review comment preview

Creating, updating, commenting, requesting review, or merging needs explicit authority.

Resolve exact identity first

Before any action, resolve:

Hosting system
Repository
Head branch
Base branch
Current commit
Existing PR or MR number, if any

Do not guess the repo from the current shell if the task can name it explicitly.

Draft workflow

Prepare a preview with:

Action:
Repo:
Head:
Base:
Title:
Body:
Linked tickets:
Exact create or update command:

The preview should include:

  • problem
  • root cause or change summary
  • proof
  • known risks or follow-up

Write workflow

Once authority is explicit:

  1. resolve repo, head, base, and current commit again
  2. create or update the PR or MR
  3. read the saved result back
  4. return the URL, state, and proof

Monitor workflow

For a watch-style use, report:

PR or MR:
Repo:
Head:
Base:
Checks:
Reviews:
Mergeability:
Next useful action:

Rules

  1. Never merge from this skill without explicit authority.
  2. Never post comments or review requests without explicit authority.
  3. Keep the repo/head/base exact. Those identities are part of the proof.
  4. Do not call a branch ready just because local tests passed if required remote checks are still pending.

Stop rules

Stop when:

  • the read result is complete
  • the draft preview is ready
  • the authorized create or update is confirmed
  • the PR or MR is waiting on human review or external checks

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.