code-review-and-quality

Conducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensions before it enters the main branch.

Quality

54%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/code-review-and-quality/SKILL.md

The canonical home for this skill is code-review-and-quality in addyosmani/agent-skills

Quality

Content

56%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This is a comprehensive code review skill that covers the domain thoroughly with a well-structured five-axis framework and clear review process. Its main weakness is significant verbosity — it reads more like a code review handbook than a concise skill file, explaining many concepts Claude already knows (review speed etiquette, disagreement handling, why dead code is bad). The actionable elements (checklist, severity labels, splitting strategies) are strong, but they're buried in substantial padding that could be trimmed by 40-50%.

Suggestions

Cut sections that teach general software engineering wisdom Claude already knows: 'Review Speed', 'Handling Disagreements', 'Common Rationalizations', and 'Honesty in Review' could be reduced to 2-3 bullet points each or removed entirely.

Move the detailed 'Change Sizing', 'Change Descriptions', and 'Dependency Discipline' sections into separate reference files to keep SKILL.md as a focused overview with pointers.

Add concrete tool commands for verification steps (e.g., specific linting commands, test runners, security scanning tools) to make the workflow more executable rather than advisory.

Add an explicit feedback loop in the review process: 'If Critical or Required issues found → return to Step 3 after author addresses feedback → re-review only changed areas.'

Dimension	Reasoning	Score
Conciseness	The skill is significantly verbose at ~350+ lines. It extensively explains concepts Claude already understands well — what code review is, why slow reviews are bad, how disagreements work, common rationalizations, and general software engineering wisdom. Much of this is general knowledge that doesn't need to be spelled out. The 'Common Rationalizations' table, 'Review Speed' section, and 'Handling Disagreements' section are largely unnecessary padding.	2 / 5
Actionability	The skill provides concrete, actionable guidance: a clear five-axis framework, severity label system with a table, a complete review checklist, specific splitting strategies, and a multi-model review pattern with example prompts. However, it lacks executable code examples for actually performing reviews (e.g., specific tool commands for running tests, linting, or security checks) and some sections remain at the level of general advice rather than specific instructions.	4 / 5
Workflow Clarity	The five-step review process (Understand Context → Review Tests → Review Implementation → Categorize Findings → Verify the Verification) is clearly sequenced and logical. The verification checklist at the end provides a validation checkpoint. However, the feedback loop for addressing issues is somewhat implicit — there's no explicit 'if issues found, iterate' step with clear re-review criteria, and the multi-model review pattern's flow diagram is helpful but lacks detail on when to loop back.	4 / 5
Progressive Disclosure	The skill references external files ('security-and-hardening', 'performance-optimization', 'references/security-checklist.md', 'references/performance-checklist.md') which is good progressive disclosure, but no bundle files are provided to support these references. The main file itself is monolithic — the Common Rationalizations table, Red Flags list, and detailed Change Sizing/Description sections could be split into reference files. The content that is inline is too much for a single SKILL.md overview.	3 / 5
	Total	13 / 20 Passed

Description

53%Scale 1-5

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 excels at defining when to use the skill with multiple explicit trigger scenarios, but falls short on specificity — it never explains what 'multi-axis' means or what concrete review actions are performed. The trigger terms are adequate but miss common synonyms like 'PR,' 'pull request,' or 'diff.' Adding concrete capabilities would significantly improve both specificity and distinctiveness.

Suggestions

List the specific review axes/dimensions (e.g., 'Evaluates security, performance, readability, test coverage, error handling, and naming conventions').

Add natural trigger terms and synonyms such as 'PR review,' 'pull request,' 'diff,' 'code feedback,' and 'review my code'.

Describe the concrete output or actions (e.g., 'Produces structured feedback with severity ratings and actionable suggestions for each dimension').

Dimension	Reasoning	Score
Specificity	The description names the domain ('code review') and mentions 'multi-axis' review across 'multiple dimensions,' but never specifies what those axes or dimensions are. No concrete actions like 'checks for security vulnerabilities, evaluates naming conventions, verifies test coverage' are listed.	2 / 5
Completeness	The 'when' is explicitly and thoroughly addressed with multiple trigger scenarios ('before merging,' 'reviewing code written by yourself, another agent, or a human,' 'assess code quality'). The 'what' is present but vague — 'multi-axis code review' doesn't explain what specific actions or outputs the skill produces.	4 / 5
Trigger Term Quality	Includes relevant terms like 'code review,' 'merging,' 'code quality,' and 'main branch,' which users might naturally say. However, it misses common synonyms and variations like 'PR review,' 'pull request,' 'diff review,' 'lint,' or 'code feedback.'	3 / 5
Distinctiveness Conflict Risk	While 'multi-axis code review' is somewhat distinctive, the description could overlap with general code quality tools, linting skills, or single-dimension review skills. The lack of specificity about what axes are covered makes it harder to distinguish from other code review or quality-related skills.	3 / 5
	Total	12 / 20 Passed

Validation

100%

Warnings & errors only

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: finlay611/agent-skills
Path: skills/code-review-and-quality/SKILL.md
Commit: 8c65303

Reviewed: ~2 hours ago

Table of Contents

Discovery Implementation Validation

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.