code-review
How to review code; a pull request, feature branch, local changes etc.
50
22%
Does it follow best practices?
Impact
100%
1.14xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./.claude/skills/code-review/SKILL.mdQuality
Discovery
22%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description is too vague about what concrete actions the skill performs during code review. It lacks an explicit 'Use when...' clause and doesn't enumerate specific capabilities like identifying bugs, checking style, analyzing diffs, or suggesting improvements. The trigger terms are decent but incomplete, missing common abbreviations like 'PR'.
Suggestions
Add specific concrete actions the skill performs, e.g., 'Analyzes code diffs, identifies bugs and security issues, checks style consistency, and suggests improvements.'
Add an explicit 'Use when...' clause, e.g., 'Use when the user asks to review a PR, check a diff, review code changes, or analyze a feature branch.'
Include common trigger term variations like 'PR', 'code review', 'diff', 'merge request', 'CR', and 'staged changes'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description says 'how to review code' which is vague. It lists contexts (pull request, feature branch, local changes) but does not describe concrete actions like 'check for bugs', 'suggest improvements', 'analyze diffs', etc. | 1 / 3 |
Completeness | The description partially addresses 'what' (review code) but is vague about specific actions. There is no explicit 'Use when...' clause or equivalent trigger guidance, which per the rubric caps completeness at 2, and the weak 'what' brings it down to 1. | 1 / 3 |
Trigger Term Quality | Includes some natural keywords users might say like 'pull request', 'feature branch', 'local changes', and 'review code', but misses common variations like 'PR', 'code review', 'diff', 'merge request', or 'changes'. | 2 / 3 |
Distinctiveness Conflict Risk | The mention of 'pull request', 'feature branch', and 'local changes' provides some specificity to code review contexts, but 'review code' is broad enough to potentially overlap with linting, testing, or general coding assistance skills. | 2 / 3 |
Total | 6 / 12 Passed |
Implementation
22%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is too vague and generic to be useful. It lists review categories Claude already knows about without providing any concrete workflow, example outputs, or specific commands for performing code reviews. The reference to CLAUDE.md and the glossary skill are the only project-specific additions of value.
Suggestions
Add a concrete workflow: e.g., 1) Fetch the diff/PR using specific commands (gh pr diff, git diff), 2) Review each file, 3) Structure feedback in a specific format with examples of good review comments.
Provide an example review output format showing how findings should be categorized and presented (e.g., severity levels, file/line references, suggested fixes).
Remove the generic bullet list of review areas (code quality, bugs, performance, security, test coverage) — Claude already knows these. Replace with project-specific conventions or non-obvious review priorities.
Add specific tool usage instructions: how to use GitHub CLI to comment on PRs, how to check out branches, how to run the project's test suite as part of the review.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Reasonably brief but the bullet list of review areas (code quality, bugs, performance, security, test coverage) is generic knowledge Claude already possesses. The lines about being 'constructive and helpful' add no value. | 2 / 3 |
Actionability | Entirely vague and abstract—no concrete commands, no example review output format, no specific steps for how to fetch a PR, diff local changes, or structure feedback. It describes what to review but not how. | 1 / 3 |
Workflow Clarity | No sequenced workflow at all. There's no guidance on how to start a review (e.g., fetch diff, read files), how to structure findings, or any validation/checkpoint steps. | 1 / 3 |
Progressive Disclosure | The skill references CLAUDE.md and the glossary skill, which is a form of progressive disclosure. However, there's no structured organization (no headings, no sections), and the content is a flat list rather than a well-organized overview. | 2 / 3 |
Total | 6 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- MetaMask/ocap-kernel
- Commit
d93dc41
- Reviewed
Table of Contents
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.