Review pr
Reviews a pull request with structured severity-based feedback. Use when asked to review a PR, asked for a code review, or given a PR number/URL.From its SKILL.md
npx -y skills add domengabrovsek/claude --skill review-prAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 14 stars14 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
2.4 KB, 561 tokens by cl100k_base, as published. Nobody here has run it
Review the pull request: $ARGUMENTS
Follow this process:
why-no-hook: review workflow guidance; each step requires reading the PR and judging code quality / intent.
- Fetch PR details: run
gh pr view $ARGUMENTS --json title,body,files,commits,additions,deletions,baseRefName,headRefName(review-time: see section note) - Read the diff: run
gh pr diff $ARGUMENTSto see all changes(review-time: see section note) - Understand intent: read the PR description, linked issues, and commit messages before reviewing code
(review-time: see section note) - Spec-conformance pass (when a spec exists): find the originating spec - issue refs in the commits (via
gh), a linked issue, or a spec under.claude/state/specs/- and check the diff against it, ideally in a parallel sub-agent so it does not pollute the main review context: (a) requirements asked for but missing or partial; (b) behaviour in the diff nobody asked for (scope creep); (c) requirements that look implemented but wrong. Quote the spec line for each finding and place it in the severity buckets below. If there is no spec, skip this pass and note it.(review-time: see section note) - Load relevant agents: based on the files changed, load the appropriate expert agents from
~/.claude/agents/for domain-specific review(review-time: see section note) - Review systematically using the checklist in @checklist.md
(review-time: see section note) - Produce structured output in this format:
(review-time: see section note)
## Summary
<1-2 sentence overall assessment>
## Verdict: APPROVE / REQUEST_CHANGES / NEEDS_DISCUSSION
<reason>
## Findings
### BLOCKER (must fix before merge)
- **[file:line]** - Description. Why it matters. Suggested fix.
### ISSUE (should fix, may approve with commitment to follow-up)
- **[file:line]** - Description. Recommendation.
### SUGGESTION (take it or leave it)
- **[file:line]** - Description. Alternative approach.
### NIT (style/preference, non-blocking)
- **[file:line]** - Description.
### PRAISE (good patterns worth highlighting)
- **[file:line]** - What's done well and why.
If no $ARGUMENTS provided, review changes in the current branch against the base branch using git diff main...HEAD.
What ships with it: 1 file
2.6 KB alongside SKILL.md
- checklist.md2.6 KB
Gives 0 of the 12 instructions most review quality skills give in 561 tokens
Counted across 1,048 of the 1,783 authors here whose files we hold, read 2026-08-07
- Ask questions one at a timein 81 of 1048, across 64 files
- Provide a recommended answer for each questionin 73 of 1048, across 50 files
- Explore the codebase instead of asking answerable questionsin 66 of 1048, across 42 files
- Resolve dependencies between decisions one-by-onein 42 of 1048, across 17 files
- Interview the user relentlessly about the planin 38 of 1048, across 13 files
- Order findings by severityin 31 of 1048
- Resolve each branch of the decision treein 27 of 1048, across 5 files
- Run a grilling sessionin 26 of 1048, across 5 files
- Update CONTEXT.md immediately when a term is resolvedin 26 of 1048, across 11 files
- Propose precise canonical terms for vague languagein 25 of 1048, across 7 files
- Create documentation files lazilyin 24 of 1048, across 5 files
- Assign severity to every findingin 24 of 1048
Said here and by no other author read
- note it if there is no spec
- load relevant domain-specific expert agents
- review systematically using the checklist
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.