code-review

Brutally honest code review assessing security, reliability, performance, and taste

Quality

57%

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 ./.opencode/skills/code-review/SKILL.md

Quality

Content

65%

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

The body is highly actionable thanks to concrete prompt and output templates, but it pays for that with significant redundancy across three restated workflows and duplicated criteria, and it keeps all detail inline rather than splitting criteria into reference files.

Suggestions

Collapse the four overlapping workflow sections (Your Task, Execution Strategy, Review Process, Begin) into one authoritative numbered sequence to remove redundancy.

Merge 'Review Criteria' and 'Subagent Review Criteria' into a single criteria section, or move the per-pass detail into a references/ file (e.g. criteria.md) that subagent prompts include.

Trim the enumerated lists of well-known issues (e.g. 'Null/undefined handling', 'Edge cases') to the few that need emphasis, trusting Claude's existing competence.

Dimension	Reasoning	Score
Conciseness	The body is mostly useful guidance, but the workflow is described three times (## Your Task, ## Execution Strategy, ## Review Process, ## Begin) and the review criteria appear twice (## Review Criteria and ## Subagent Review Criteria), so it 'could be tightened' as the score-2 anchor describes; it also restates security/reliability concepts Claude already knows.	2 / 3
Actionability	It provides a copy-paste-ready subagent prompt template (with Files/Context/Focus placeholders and a structured return format) and a full output-format template with severity sections and example issue blocks, satisfying the score-3 anchor 'Fully executable code/commands; specific examples; copy-paste ready'.	3 / 3
Workflow Clarity	Steps are sequenced (scope -> gather -> dispatch -> aggregate -> write), but the same sequence is restated in three sections with no single authoritative flow, creating the 'sequence present but checkpoints missing or implicit' condition of the score-2 anchor; the operation is non-destructive so a validation loop is not strictly required.	2 / 3
Progressive Disclosure	No bundle files exist and the entire 350-line skill lives inline in SKILL.md; sections are well-headed (so it is not a wall of text), but the detailed review criteria and subagent criteria are inline content that 'should be separate' per the score-2 anchor, fitting 'Some structure but could be better organized'.	2 / 3
	Total	9 / 12 Passed

Description

50%

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 third-person and conveys a clear, distinctive purpose (code review across four dimensions), but it omits any 'Use when...' trigger guidance and offers only one action verb, leaving it competent yet incomplete.

Suggestions

Add an explicit trigger clause, e.g. 'Use when the user asks to review code, audit a PR, or get brutally honest feedback on a file or diff.'

Expand the action verbs beyond 'assessing' to concrete outputs, e.g. 'Reviews code and writes an actionable report grading security, reliability, performance, and taste.'

Dimension	Reasoning	Score
Specificity	The phrase 'code review assessing security, reliability, performance, and taste' names the domain and several review dimensions, but uses a single action verb ('assessing') rather than enumerating multiple distinct concrete actions, matching the score-2 anchor 'Names domain and some actions, but not comprehensive'.	2 / 3
Completeness	It clearly states what the skill does ('code review assessing...') but provides no 'Use when...' clause or equivalent explicit trigger guidance, so per the judging guidelines completeness is capped at 2 ('Has what, but when is missing or only implied').	2 / 3
Trigger Term Quality	'code review' is a natural term a user would say, and 'security, reliability, performance, taste' are relevant, but coverage of common variations ('review my code', 'code audit', 'look over this PR') is thin, fitting the score-2 anchor 'Some relevant keywords but missing common variations'.	2 / 3
Distinctiveness Conflict Risk	'code review' is a recognizable niche, yet without explicit trigger guidance it could still overlap with general review or refactoring skills, matching the score-2 anchor 'Somewhat specific but could still overlap with similar skills'.	2 / 3
	Total	8 / 12 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: LibPDF-js/core
Path: .opencode/skills/code-review/SKILL.md
Commit: 36bf165

Reviewed: 2 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.