review
Code review with semantic diffs, expert routing, and auto-TaskCreate. Triggers on: code review, review changes, check code, review PR, security audit.
81
76%
Does it follow best practices?
Impact
85%
1.34xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./data/skills-md/0xdarkmatter/claude-mods/review/SKILL.mdQuality
Discovery
89%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 strong on trigger terms and completeness, with explicit trigger guidance and natural keywords. Its main weakness is that the capability descriptions read more like feature bullet points ('semantic diffs', 'expert routing', 'auto-TaskCreate') rather than concrete actions, which reduces specificity. 'auto-TaskCreate' in particular is jargon that doesn't clearly communicate what it does.
Suggestions
Replace feature labels with concrete action descriptions, e.g., 'Performs code review by analyzing semantic diffs, routing findings to domain experts, and automatically creating follow-up tasks' instead of listing feature names.
Clarify 'auto-TaskCreate' with plain language, e.g., 'automatically creates follow-up tasks for identified issues'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (code review) and mentions some features like 'semantic diffs', 'expert routing', and 'auto-TaskCreate', but these are more like feature labels than concrete actions. It doesn't describe what specific actions are performed (e.g., 'analyzes diffs for security vulnerabilities, suggests improvements, routes to domain experts'). | 2 / 3 |
Completeness | Answers both 'what' (code review with semantic diffs, expert routing, auto-TaskCreate) and 'when' (explicit 'Triggers on:' clause with specific trigger terms). The trigger guidance is explicit and clear. | 3 / 3 |
Trigger Term Quality | Explicitly lists natural trigger terms: 'code review', 'review changes', 'check code', 'review PR', 'security audit'. These are terms users would naturally say when requesting code review functionality, with good coverage of common variations. | 3 / 3 |
Distinctiveness Conflict Risk | The combination of 'semantic diffs', 'expert routing', and 'auto-TaskCreate' creates a distinct niche. The trigger terms are specific to code review workflows and unlikely to conflict with general coding or document skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
62%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is highly actionable with excellent workflow clarity and concrete, executable examples throughout. However, it is significantly over-engineered and verbose for a skill file—much of the content (ASCII architecture diagrams, extensive routing tables, severity definitions, multiple mode tables) inflates token cost without proportional value. The progressive disclosure is partially implemented with one external reference but the main file remains a wall of content that should be decomposed.
Suggestions
Reduce the main SKILL.md to ~100 lines by moving the expert routing table, severity system, focus/depth modes, and advanced flags into separate reference files (e.g., routing.md, modes.md, advanced.md)
Remove the ASCII architecture diagram—the execution steps already convey the workflow clearly and the diagram is redundant
Trim the example review output to a minimal template rather than a full worked example; Claude can generate appropriate formatting from a brief schema
Provide the referenced framework-checks.md bundle file or remove the reference to avoid a broken link
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~300+ lines. It over-explains architecture with an ASCII flowchart, includes extensive tables for routing/severity/focus/depth modes, and provides lengthy example outputs that Claude could generate on its own. Much of this (severity definitions, expert routing tables, CLI tool descriptions) is padding that doesn't add actionable value proportional to its token cost. | 1 / 3 |
Actionability | The skill provides fully executable bash commands for each step, concrete code examples with diff blocks, specific file patterns for routing, and copy-paste ready CI/CD integration YAML. The guidance is specific and concrete throughout. | 3 / 3 |
Workflow Clarity | The 6-step workflow is clearly sequenced with explicit validation checkpoints. Step 1 handles scope determination with fallbacks, Step 5 produces structured output, and Step 6 includes task creation for critical issues with dependency linking and status updates. The feedback loop for --fix mode includes confirmation prompts before applying changes. | 3 / 3 |
Progressive Disclosure | The skill references `framework-checks.md` for detailed framework-specific patterns, which is good progressive disclosure. However, the main file is monolithic with extensive inline content (routing tables, severity systems, focus modes, depth modes, advanced flags) that could be split into separate reference files. No bundle files are provided to support the reference. | 2 / 3 |
Total | 9 / 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
- NeverSight/skills_feed
- Commit
6980586
- 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.