code-review
Code review practices with technical rigor and verification gates. Use for receiving feedback, requesting code-reviewer subagent reviews, or preventing false completion claims in pull requests.
81
72%
Does it follow best practices?
Impact
94%
1.06xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/code-review/skills/code-review/SKILL.mdQuality
Discovery
67%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 adequately covers both what the skill does and when to use it, earning good marks for completeness. However, the capabilities described are somewhat abstract rather than listing concrete actions, and the trigger terms could be expanded to include more natural user language variations. The skill's niche around verification gates and subagent reviews provides some distinctiveness but could be clearer.
Suggestions
Add more concrete actions such as 'enforce test coverage checks, validate CI status, verify no unresolved comments' to improve specificity.
Include common natural trigger term variations like 'PR review', 'review my code', 'merge request', 'diff feedback' to improve discoverability.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names the domain (code review) and mentions some actions like 'receiving feedback', 'requesting code-reviewer subagent reviews', and 'preventing false completion claims', but these are somewhat abstract rather than concrete technical actions. It doesn't list specific operations like 'check for test coverage', 'validate linting', or 'verify CI passes'. | 2 / 3 |
Completeness | The description answers both 'what' (code review practices with technical rigor and verification gates) and 'when' (Use for receiving feedback, requesting code-reviewer subagent reviews, or preventing false completion claims in pull requests), with explicit trigger scenarios. | 3 / 3 |
Trigger Term Quality | Includes some relevant terms like 'code review', 'pull requests', 'feedback', and 'subagent reviews', but misses common natural variations users might say such as 'PR review', 'review my code', 'code feedback', 'merge request', or 'diff review'. The term 'verification gates' is somewhat jargon-heavy. | 2 / 3 |
Distinctiveness Conflict Risk | The mention of 'code review' and 'pull requests' could overlap with general git/GitHub skills or CI/CD skills. The 'subagent reviews' and 'verification gates' aspects add some distinctiveness, but 'code review practices' is broad enough to potentially conflict with other development workflow skills. | 2 / 3 |
Total | 9 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured skill with strong actionability and workflow clarity—concrete commands, explicit verification gates, and clear anti-patterns make it highly usable. The main weaknesses are moderate redundancy between the overview/summary sections and the protocol sections within the same file, and the inability to verify that referenced bundle files exist. The decision tree and Iron Law are particularly effective anchoring devices.
Suggestions
Reduce redundancy by trimming the protocol summaries in the main file—since full protocols are in reference files, the SKILL.md sections could be shorter pointers rather than near-complete summaries.
Provide the referenced bundle files (references/code-review-reception.md, etc.) so the progressive disclosure structure can be fully validated and the main file can be leaner.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably efficient but has some redundancy—the overview section, the quick decision tree, and the protocol sections repeat similar information. The trigger lists and protocol summaries could be tighter, though it avoids explaining concepts Claude already knows. | 2 / 3 |
Actionability | Provides concrete bash commands for verification, specific git commands for requesting reviews, explicit anti-patterns (performative agreement examples), a clear decision tree, and step-by-step protocols. The guidance is specific and directly executable. | 3 / 3 |
Workflow Clarity | Multi-step processes are clearly sequenced (READ → UNDERSTAND → VERIFY → EVALUATE → RESPOND → IMPLEMENT; IDENTIFY → RUN → READ → VERIFY → CLAIM). Verification gates are explicit with clear checkpoints, feedback loops (fix Critical immediately, re-verify), and red flags for stopping. The Iron Law provides an unambiguous validation requirement. | 3 / 3 |
Progressive Disclosure | The skill references three separate files (code-review-reception.md, requesting-code-review.md, verification-before-completion.md) with clear signaling, which is good structure. However, no bundle files were provided, so we cannot confirm these references resolve correctly. Additionally, the SKILL.md itself includes substantial protocol detail that partially duplicates what the reference files presumably contain, blurring the line between overview and detail. | 2 / 3 |
Total | 10 / 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
- secondsky/claude-skills
- Commit
5e92b71
- 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.