Name: work-with-pr
Rating: 69.6 (1 reviews)
Author: code-yeongyu

work-with-pr

Full PR lifecycle in a fresh task-owned git worktree: implement via the ulw-loop skill with mandatory evidence-bound manual QA → reviewer-readable English PR → verification loop (CI + Cubic, where Cubic is skipped only when its quota is exhausted) → merge by default → worktree cleanup. Decomposes one task into the smallest atomic, independently-mergeable PRs and builds the independent ones concurrently via one worktree per PR driven by parallel subagents or a team. Unbounded loop: any failing gate sends you back to fix-and-re-QA inside that PR's worktree. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'split into atomic PRs', 'parallel PRs', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.

Quality

84%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

The canonical home for this skill is work-with-pr in code-yeongyu/oh-my-openagent

Quality

Content

77%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, clearly sequenced PR-lifecycle workflow with strong validation feedback loops. Its main weaknesses are repetition of the Cubic-skip rule across sections and a monolithic single-file structure with no progressive disclosure into reference bundles.

Suggestions

State the Cubic-quota-skip rule once in Gate B and reference it from the architecture block and Anti-Patterns table instead of restating it four times.

Move the full PR-body template and the detailed Cubic review-parsing/monitor snippets into a references/ file (e.g. references/pr-template.md), keeping SKILL.md an overview with one-level-deep links.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and free of basic-concept padding, but the Cubic-quota-skip rule is restated four times (architecture block, Phase 3 intro, Gate B section, Anti-Patterns table) and the gate ordering is repeated, which is unnecessary re-explanation that could be tightened.	3 / 5
Actionability	Fully executable, copy-paste-ready guidance throughout: concrete gh/git worktree commands, monitor() subscription snippets, and a complete PR-body EOF template covering the common cases.	5 / 5
Workflow Clarity	Clear Phase 0→4 sequence with explicit validation checkpoints (Gate A CI, Gate B Cubic), a documented unbounded feedback loop (failing gate → Phase 1 fix + re-QA → re-verify), and an iteration-discipline checklist.	5 / 5
Progressive Disclosure	Well-organized into phases with section headers, but it is a ~400-line monolithic SKILL.md with no bundle files in references/scripts/assets; content that could live one level deep (the full PR-body template, the Cubic review-parsing logic) is inlined rather than split.	3 / 5
	Total	16 / 20 Passed

Description

91%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description that clearly states the full PR lifecycle and when to use it, with a rich natural trigger list. Minor deductions for a second-person clause and a broad catch-all trigger that risks overlap with sibling implementation skills.

Suggestions

Rewrite 'sends you back to fix-and-re-QA' in third person (e.g., 'routes back to fix-and-re-QA') to satisfy the voice rule and recover the specificity point.

Tighten the 'even when user just says implement X' catch-all to require a clearer PR-delivery signal to reduce conflict with general implementation skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete lifecycle actions ('worktree setup', 'implement via the ulw-loop skill with mandatory evidence-bound manual QA', 'reviewer-readable English PR', 'verification loop (CI + Cubic)', 'merge by default', 'worktree cleanup') with comprehensive coverage, but the second-person phrase 'sends you back to fix-and-re-QA' triggers the -1 voice penalty.	4 / 5
Completeness	Explicitly answers both what ('Full PR lifecycle in a fresh task-owned git worktree...') and when ('Use whenever implementation work needs to land as a PR' plus concrete trigger phrases).	5 / 5
Trigger Term Quality	Comprehensive natural trigger list including 'create a PR', 'implement and PR', 'land this as a PR', 'split into atomic PRs', 'parallel PRs', 'PR workflow', and 'implement end to end' — phrases a user would naturally say.	5 / 5
Distinctiveness Conflict Risk	Has a clear niche (PR lifecycle orchestration with the CI+Cubic verification loop) with mostly distinct triggers, but the catch-all 'even when user just says "implement X"' creates minor overlap risk with general implementation skills like ulw-loop.	4 / 5
	Total	18 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: code-yeongyu/oh-my-openagent
Path: .agents/skills/work-with-pr/SKILL.md
Commit: 6108673

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.