Menu
Review a pull request using Lean waste categories and structured severity ratings
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Health check procedures D1–D14 for the Audit agent — structural validation, attention budget, version checks, workspace integrity, and static audit
Configure and manage Model Context Protocol servers for external tool access
Review a UI for accessibility — WCAG 2.1 AA compliance, semantic HTML, ARIA usage, keyboard navigation, focus management, colour contrast, and screen reader compatibility
Design or review a REST or GraphQL API — resource modeling, versioning strategy, error contract, OpenAPI/schema-first workflow, and security baseline
Generate a CHANGELOG.md entry from staged changes, a commit range, or a PR diff — following Keep a Changelog format with conventional commit classification
Set up and audit environment variable management — create .env.example, add startup validation, separate secrets from config, and document every variable
| name | lean-pr-review |
| description | Review a pull request using Lean waste categories and structured severity ratings |
| compatibility | >=1.4 |
Skill metadata: version "1.1"; license MIT; tags [review, pull-request, lean, kaizen, code-review]; compatibility ">=1.4"; recommended tools [codebase, githubRepo].
Perform a structured pull request review using §5 Review Mode conventions and §6 waste categories.
Get the diff — Read PR diff or changed files. Locally: git diff main...HEAD.
Scan each changed file — Read full files (not just hunks) for context.
Classify each finding — [severity] | [file:line] | [waste category] | [description]
Severity: critical (blocks merge), major (fix before merge), minor (nice to fix), advisory (informational).
Waste categories (§6) — full list W1–W16; most common in PR review:
| Code | Name | Typical PR signal |
|---|---|---|
| W1 | Overproduction | Dead code, unused exports, features not yet needed |
| W2 | Waiting | Blocking sync calls, missing timeouts |
| W3 | Transport | Unnecessary data copying, prop drilling 3+ levels |
| W4 | Over-processing | Abstraction for its own sake, premature generalisation |
| W5 | Inventory | Large WIP; changes that could be split into smaller PRs |
| W6 | Motion | Logic scattered across many files without justification |
| W7 | Defects | Bugs, type errors, missing error handling, test failures |
| W8 | Unused talent | Missing tests, missing automation, repetitive manual patterns |
| W11 | Hallucination rework | Phantom API usage, methods that don't exist, incorrect assumptions |
| W14 | Model-task mismatch | Overly complex solution to a trivial problem |
For W9–W10, W12–W13, W15–W16 definitions, see §6 of .github/copilot-instructions.md.
Check test coverage — New/changed behaviour must have tests. Flag untested paths as major | W7.
Check baselines — §2 baselines: file LOC limits, dependency budget, zero type errors.
Produce the report — Format as:
## PR Review — <PR title or branch name>
### Summary
<1–2 sentence overview of the changes and their quality>
### Findings (<N> total: <critical> critical, <major> major, <minor> minor, <advisory> advisory)
#### Critical
- [critical] | [file:line] | [W7] | <description>
#### Major
- [major] | [file:line] | [W4] | <description>
#### Minor
- [minor] | [file:line] | [W1] | <description>
#### Advisory
- [advisory] | [file:line] | [W8] | <description>
### Verdict
<APPROVE / REQUEST CHANGES / COMMENT>
Wait — Do not apply fixes. Present the report and wait for the user to decide what to address.