code-review

Review code for actionable defects. Correctness is the core; performance and security are optional sub-cases of the same engine. Anchors on agreements between participants across a boundary, forces a violating execution, and refutes every candidate before reporting. Use when asked to review changes, find bugs, audit a codebase, or check whether a fix is safe.

Quality

84%

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

Quality

Content

77%Weight 40%Scale 1-5

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

A well-structured, highly actionable instruction-only methodology skill with clear sequencing, explicit validation gates, and clean one-level-deep progressive disclosure. The main weakness is moderate verbosity in the philosophical framing and the absence of a concrete worked example of a completed finding report.

Suggestions

Tighten the Purpose and 'Absorption is not prevention' passages to their operational core to reduce token cost without losing the methodology.

Add one short worked example showing the report template populated with a real cross-boundary finding to lift actionability from concrete-template to demonstrated.

Trim repeated re-emphasis of the cross-boundary theme across Structure, Correctness, and Parallelism to a single statement plus back-references.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient methodology guidance that does not explain basic concepts Claude already knows, but the rhetorical framing in the Purpose section and extended philosophical passages (e.g. 'Absorption is not prevention', 'Its central bet') could be tightened without losing substance.	3 / 5
Actionability	Provides concrete, copy-paste-ready templates (the working-notes block, the five-element refutation checklist, the full report skeleton) and an explicit procedure skeleton, but lacks a worked example showing the report format filled in with a real finding.	4 / 5
Workflow Clarity	The shared engine lays out an explicit sequenced skeleton ('Map a flow → identify an obligation → ... → report'), the refutation discipline is an explicit validation gate with a Keep/Drop/Unresolved feedback rule, and the five required elements plus the parallelism coordinator/worker steps serve as checklists.	5 / 5
Progressive Disclosure	The body is a clear overview that splits sub-case detail into three real one-level-deep reference files (correctness-taxonomy.md, performance-review.md, security-review.md), signaled in a table and inline with explicit instruction to read each only when its sub-case is selected.	5 / 5
	Total	17 / 20 Passed

Description

83%Weight 40%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.

A strong description that concretely states both capability and trigger conditions in third person, with a distinct methodological niche. It could be sharper by adding common synonyms and acknowledging the disambiguation from the bundled /code-review directly in the description.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — 'Review code for actionable defects', 'Anchors on agreements between participants across a boundary', 'forces a violating execution', 'refutes every candidate before reporting' — but they are methodological rather than tangible operations, leaving minor coverage gaps versus a fully comprehensive list.	4 / 5
Completeness	It explicitly states what the skill does ('Review code for actionable defects...refutes every candidate before reporting') and when to use it ('Use when asked to review changes, find bugs, audit a codebase, or check whether a fix is safe'), with concrete trigger phrases for both.	5 / 5
Trigger Term Quality	The 'Use when asked to review changes, find bugs, audit a codebase, or check whether a fix is safe' clause supplies several natural phrases a user would say, though common synonyms like 'inspect', 'lint', or the bare phrase 'code review' are absent.	4 / 5
Distinctiveness Conflict Risk	The agreement/refutation methodology is a clear niche, but the trigger terms overlap with generic code-review tooling (the body itself notes Claude Code ships a bundled /code-review), so minor conflict risk with a closely related skill remains.	4 / 5
	Total	17 / 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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: phuryn/pm-skills
Path: pm-ai-shipping/skills/code-review/SKILL.md
Commit: 8607e3b

Reviewed: 6 days 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.