Agent skill
pr-review-camino-school-lanttern
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/pr-review-camino-school-lanttern
SKILL.md
Your task
You are an expert code reviewer specializing in Phoenix/Elixir applications, with deep knowledge of the Lanttern educational assessment platform's codebase, patterns, and quality standards.
The user has invoked /pr-review with a PR number (e.g. #123 or just 123). Extract that number from their input, then follow the setup steps below before reviewing.
Display the review in the terminal only — do NOT post comments, create GitHub issues, or push/write to any remote repository.
Step 1 — Gather PR metadata
Run the following via Bash (replace <PR> with the number from user input):
gh pr view <PR> --json number,title,url,state,headRefName,baseRefName,body,closingIssuesReferences,reviews,comments,reviewThreads
This gives you:
headRefName— the PR branchbaseRefName— the target branch (may not bemain)body— PR descriptionclosingIssuesReferences— GitHub issues this PR closes (use these as issue context, same as--issuesin/review)reviews/comments/reviewThreads— full history for PR History Awareness
If closingIssuesReferences contains issues, fetch each one for full context:
gh issue view <number> --json number,title,body
Step 2 — Sync branches locally
Fetch both branches so diffs and history are accurate:
git fetch origin <headRefName> <baseRefName>
Step 3 — Build diff context
Using the fetched refs, run these via Bash:
git diff origin/<baseRefName>...origin/<headRefName>
git diff origin/<baseRefName>...origin/<headRefName> --name-only
git diff origin/<baseRefName>...origin/<headRefName> --stat -- 'lib/**'
git log origin/<baseRefName>..origin/<headRefName> --oneline
Step 4 — PR History Awareness
Using the data from Step 1, before reviewing:
- Previous change requests: If reviewers previously requested changes, verify those have been addressed. Flag any that remain unresolved.
- Prior comments: Take prior inline review comments into account — do not repeat feedback that was already acknowledged and resolved.
- Review state: Note if the PR was previously approved, dismissed, or had change requests, and consider this in your overall assessment.
- Unresolved threads: Highlight any review threads that appear unresolved based on the PR data.
Step 5 — Apply the review checklist
Read the shared checklist file at .claude/skills/review-checklist.md and apply it in full.
Framing guidance for this skill: The overall assessment in the Overview section should be framed as one of:
- "Approved"
- "Approved with minor suggestions"
- "Needs changes before merge"
IMPORTANT: Output the review to the terminal only. You may use gh in read-only mode to fetch PR data (reviews, comments, threads, issues). Do NOT use gh to post comments, request changes, approve, or otherwise write to the remote repository. Do not use git push or any other write operation against the remote.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?