code-review

Use when code has been written and needs validation before committing, or when the user asks for a code review or security check.

Quality

63%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./plugins/claude-code/skills/code-review/SKILL.md

Quality

Content

62%

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

The content shines on workflow clarity with a well-sequenced, validated review process, but it is held back by redundant severity restatements, motivational padding, non-executable pre-load commands, and an all-inline single-file structure.

Suggestions

Remove redundancy: the "Review Checks" severity list duplicates the example report, and "Red Flags" / "Common Rationalizations" are motivational padding — consolidate or cut them.

Replace the non-executable `Read:` lines in the bash block with real commands or a clearly labeled instruction so the pre-load step is copy-paste ready.

Move the detailed severity checklist and rationalizations into a reference file (e.g. reference/REVIEW-CHECKS.md) and link to it from SKILL.md to enable progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient but carries redundancy and padding: the "Review Checks" severity taxonomy restates the example report, and the "Red Flags" / "Common Rationalizations" sections are motivational rather than technical guidance, fitting the anchor-2 "could be tightened" case.	2 / 3
Actionability	Concrete slash commands (`/code-review path/to/file.ts`) and example diff fixes are present, but the pre-load step uses non-executable `Read:` pseudo-commands inside a bash block and the core review logic is delegated to an external subagent rather than given as executable guidance.	2 / 3
Workflow Clarity	The four-step process (Pre-load → Invoke → Analyze → Take action) is clearly sequenced with an explicit validation feedback loop ("Re-run /code-review to verify", "Proceed only when clean") and conditional branching, matching the anchor-3 checklist-with-checkpoints example.	3 / 3
Progressive Disclosure	It is a single monolithic file with well-organized sections but no progressive disclosure: detailed severity checklists and rationalizations are inline rather than split into referenced files, and there are no bundle files to link out to.	2 / 3
	Total	9 / 12 Passed

Description

65%

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 has strong, natural trigger language but leads with only the "when" and never explicitly states what the skill does, leaving completeness and specificity at the mid band. Tightening the triggers and adding an explicit capabilities clause would raise it.

Suggestions

Add an explicit "what" clause before the trigger, e.g. "Reviews written code for security, correctness, and quality. Use when...", so the description answers both what and when.

List concrete review actions (e.g. scans for injection, auth bypass, missing error handling) instead of generic categories to lift specificity.

Narrow the generic "needs validation before committing" trigger to reduce overlap with test-generation and security-audit skills.

Dimension	Reasoning	Score
Specificity	The description names the domain and a few actions ("needs validation before committing", "code review", "security check") but these are generic review categories rather than multiple specific concrete actions like the anchor-3 example listing extract/fill/merge.	2 / 3
Completeness	The text is entirely a "when" clause ("Use when...") with no standalone explicit statement of what the skill does; the "what" is only implied through the triggers, so it mirrors the anchor-2 case of one half explicit and the other implied.	2 / 3
Trigger Term Quality	"code review" and "security check" are exactly the natural phrases a user would say, and "validation before committing" adds a common variant, giving good coverage of real trigger language.	3 / 3
Distinctiveness Conflict Risk	"code review" signals a clear niche, but "needs validation before committing" and "security check" are generic enough to overlap with test-generation or dedicated security-audit skills, matching the anchor-2 "could still overlap" case.	2 / 3
	Total	9 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: darrenhinde/OpenAgentsControl
Path: plugins/claude-code/skills/code-review/SKILL.md
Commit: 37ca233

Reviewed: 3 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.