Home

code-review-skill Wiki

code-review-skill provides two portable Code Review Agent Skills that share one review standard: one for local, not-yet-PR'd changes, one for an existing GitHub Pull Request. Each Skill is packaged as a standalone Agent Skills archive and runs on any compatible runtime (Claude Code, Codex, Cursor, OpenCode, …) — see the repository README.

This Wiki is explanatory and navigational only. It summarizes and links to the canonical repository files; it never defines review behavior itself. Where anything here appears to disagree with a file in the repository, the repository file wins — see Repository-Documentation-Map.

Who this is for

• Engineers who want to review their own local changes before opening a PR, or review someone else's open PR.
• Contributors who want to understand the repository's own development rules before opening a pull request against it.
• Evaluators comparing this project's governance model against other AI code-review products.

The two Skills, in one sentence each

Skill	Reviews	Delivers
local-code-review	your local Git delta — committed, staged, unstaged, untracked	one structured report back to the caller
github-pr-review	an existing GitHub Pull Request	a report (passive), or inline comments + an Approve/Request Changes decision (active)

Not sure which one you need? See Choosing-a-Review-Skill.

Where to start

• New to the project → Getting-Started
• Want the mental model before diving into policy files → How-the-Review-Model-Works and Architecture-Overview
• Want to know when a review goes deeper than its base pass → Specialist-Depth-Review
• Using local-code-review → Local-Code-Review-Guide
• Using github-pr-review → GitHub-PR-Review-Guide
• Want a per-feature at-a-glance page (runtime validation, requirement coverage, delta re-review, GitHub publication, human-style output, …) → Feature-Pages
• Contributing to this repository → Development-Workflow
• Understanding this repository's own review-quality benchmark and its PR-time selection/gating pipeline → Benchmark-Testing
• "Where's the canonical source for X?" → Repository-Documentation-Map
• Comparing against CodeRabbit, Qodo, Greptile, Copilot, Cursor, Amazon Q → AI-Code-Review-Landscape
• Quick answers → FAQ

Where canonical rules actually live

This Wiki never redefines behavior. The repository owns all normative content:

• Repository-development rules (branching, PRs, merges, docs style): AGENTS.md and policies/
• Shared review semantics (scope, severity, evidence, context): shared/policies/
• Per-Skill behavior: each Skill's own SKILL.md, skills/local-code-review/, skills/github-pr-review/
• Architecture: docs/ARCHITECTURE.md

Full ownership map: Repository-Documentation-Map.

---

Wiki Page Map

Page	What it covers
Home	Wiki entry point: what the two Skills are and where to go next.
Getting-Started	First steps to package and run either Skill.
Choosing-a-Review-Skill	How to decide between local-code-review and github-pr-review.
How-the-Review-Model-Works	Concept-level walkthrough of the review pipeline shared by both Skills.
Architecture-Overview	Component-level map of how the repository's pieces fit together.
Specialist-Depth-Review	When and how a review goes deeper than its base pass.
Local-Code-Review-Guide	Usage guide for reviewing local, not-yet-PR'd Git changes.
GitHub-PR-Review-Guide	Usage guide for reviewing an existing GitHub Pull Request.
Review-Context-and-Existing-Evidence	How optional requirements/context and prior review evidence are supplied.
Feature-Pages	Index of per-feature capability pages and their shared template.
Feature-Delta-Re-Review	How a re-review scopes itself to only what changed since the last pass.
Feature-GitHub-Publication-and-Authorization	What a PR review publishes to GitHub and under what authority.
Feature-Human-Style-Review-Output	The senior-engineer-voice prose summary output option.
Feature-Requirement-Coverage	How supplied requirements are checked and reported as covered or not.
Feature-Runtime-Validation-Evidence	How a review can execute commands or reproductions as evidence.
Severity-and-Decision-Model	The P0/P1/P2 severity levels and how the merge decision is derived.
Packaging-and-Portability	How the Skills go from source repository to a standalone archive.
Development-Workflow	Contributor guide for working on this repository itself.
Benchmark-Testing	Orientation to this repository's own review-quality benchmark pipeline.
Repository-Documentation-Map	Where each concern's canonical, authoritative source lives.
AI-Code-Review-Landscape	Neutral snapshot comparing this project against other AI code-review products.
FAQ	Short, practical answers to common questions.