zf-reviewer-ponytail
Skill2 starsChanged 40 days agoApache-2.0View on GitHub
Pull Request review gatekeeping playbook for zf-reviewer applying the 7-Rung Ladder of Laziness.
In this repository5 other files it gives its agents.
- openwikiSkill
- ponytailSkill
- zerofactory-orchestrationSkill
- zf-builder-ponytailSkill
- zf-orchestrator-ponytailSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 2 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 26
When reviewing a PR branch diff (task/<task_id>):
- Distinctive: line 30
Leftover Artifacts: Ensure no commented-out code, temporary debug logs, or unused variables were committed.
- Very specific: line 33
Check package.json, poetry.lock, requirements.txt, etc.
- Distinctive: line 34
If a new external package was added, verify whether:
- Very specific: line 37
If unnecessary, request removal: *"Reject new dependency <pkg>.
- Very specific: line 47
When reviewing on GitHub (gh pr review --comment), structure feedback with reference to the Ponytail ladder:
≈ “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.
Files
ZF Reviewer Ponytail — PR Review Gatekeeping Playbook
"The best code is the code you never wrote."
This skill equips zf-reviewer with the Ponytail review rubric: actively catching over-engineering, diff bloat, and redundant dependencies across all Pull Request reviews (starting immediately in Round 1: Initial Review and verifying fixes in Round 2: Re-Verification).
The core decision framework is defined in the shared 7-Rung Ladder of Laziness.
Reviewer's Ponytail Checklist
When reviewing a PR branch diff (task/<task_id>):
1. Diff Scope & Hygiene (Rung 7: Minimum Viable Solution)
- Zero Drive-By Churn: Did the author reformat unrelated code, reorder imports, or touch files outside the task scope? Reject diff bloat immediately in Round 1.
- Leftover Artifacts: Ensure no commented-out code, temporary debug logs, or unused variables were committed.
2. Dependency Veto (Rung 3 & 5: Standard Library & Installed Dependencies)
- Check
package.json,poetry.lock,requirements.txt, etc. - If a new external package was added, verify whether:
- A standard library module could have handled it (Rung 3).
- An already-installed library could have solved it (Rung 5).
- If unnecessary, request removal: "Reject new dependency
<pkg>. Standard library<module>handles this in 3 lines."
3. Abstraction & Simplicity Check (Rungs 1 & 6: YAGNI & Simple Functions)
- Did the author create a class hierarchy, factory, or adapter where a single pure function suffices?
- Are there speculative parameters or "future-proofing" flags not requested in the task? Request their deletion.
Actionable Review Feedback Format
When reviewing on GitHub (gh pr review --comment), structure feedback with reference to the Ponytail ladder:
[AI:zf-reviewer] [Reviewer Feedback] Round 1: Ponytail Gatekeeping & Correctness
1. **Unnecessary Dependency (Rung 3/5)**:
- File: `src/processor.py:12`
- Problem: Added `requests` when the repo already standardizes on native `urllib.request` / `httpx`.
- Action: Remove `requests` from `requirements.txt` and use native stdlib.
2. **Diff Bloat & Drive-by Churn (Rung 7)**:
- File: `src/utils.py:1-40`
- Problem: Re-ordered imports and reformatted comments unrelated to the task.
- Action: Revert changes to `src/utils.py` to keep the PR focused on the task scope.
3. **Over-Engineered Factory (Rung 1/6)**:
- File: `src/auth.py:45-80`
- Problem: Introduced `AbstractTokenValidatorFactory` with only one concrete implementation.
- Action: Collapse into a single `validate_token(token)` function.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
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.