agent-reviewer

Agent skill for reviewer - invoke with $agent-reviewer

1.15x

Quality

27%

Does it follow best practices?

Impact

81%

1.15x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/agent-reviewer/SKILL.md

Quality

Content

27%

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

The content is a long, monolithic set of generic code-review checklists and teaching examples that re-explain well-known concepts, with no progressive disclosure to external files and no explicit validation/feedback checkpoints in its workflow. It offers some concrete examples but is too padded and not split for navigation.

Suggestions

Cut generic explanations of SOLID/DRY/KISS and N+1 queries that Claude already knows; keep only skill-specific review checklists and project-specific thresholds.

Add explicit validation/feedback steps to the review process, e.g. 'Run npm run lint/test/security-scan, then triage failures and re-check before reporting.'

Move large reference material (full MCP tool-integration blocks, example templates) into a references/ file and link to it from a concise overview so the body is navigable.

Dimension	Reasoning	Score
Conciseness	The body is verbose and explains concepts Claude already knows (general SOLID/DRY/KISS principles, what an N+1 query is, basic SQL-injection examples), padded with commentary lines like 'Reviews are learning opportunities' and redundant checklists that don't earn their tokens.	1 / 3
Actionability	It provides some concrete, runnable example code (SQL parameterization, N+1 fix) and an automated-checks command block, but many examples are illustrative skeletons ('function calculateDiscount(entity, rules) { ... }') rather than copy-paste-ready, executable guidance.	2 / 3
Workflow Clarity	A review process is sequenced (Functionality, Security, Performance, Code Quality, Maintainability) with a numbered list, but there are no explicit validation checkpoints or validate→fix→retry feedback loops for the review steps themselves; per the rubric, missing validation/feedback loops for a multi-step process caps this at 2.	2 / 3
Progressive Disclosure	The body is a monolithic wall of text with no external references and no bundle files (references/scripts/assets directories absent); the MCP tool integration and large example blocks are inlined rather than split into separate reference files.	1 / 3
	Total	6 / 12 Passed

Description

27%

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 frontmatter description is essentially a placeholder ('Agent skill for reviewer - invoke with $agent-reviewer') that fails to state what the skill does or when to use it, and lacks natural trigger keywords. The actual capabilities are buried in a separate YAML block in the body rather than the description.

Suggestions

Rewrite the frontmatter description to state concrete actions in third person, e.g. 'Reviews code for quality, security, performance, and maintainability; suggests fixes with severity rankings.'

Add an explicit 'Use when...' trigger clause covering natural terms users would say, e.g. 'Use when the user asks for a code review, security audit, or to check code quality.'

Remove the invocation-token instruction ('$agent-reviewer') from the description and surface real capability details instead of generic 'Agent skill for reviewer'.

Dimension	Reasoning	Score
Specificity	The description's effective text ('Agent skill for reviewer - invoke with $agent-reviewer') names concrete capability (review) and a concrete invocation action, and the broader skill declares multiple specific concrete review actions via the capabilities list (code_review, security_audit, performance_analysis, documentation_review).	3 / 3
Completeness	The frontmatter description answers neither 'what' clearly (only the generic 'Agent skill for reviewer') nor 'when' to use it; the judging guidelines cap completeness at 2 when a 'Use when...' clause is missing, but here both what and when are effectively absent/weak.	1 / 3
Trigger Term Quality	No natural user-facing trigger keywords are present; the only phrase is the jargony invocation token 'invoke with $agent-reviewer', which a user would not naturally say to describe needing this skill.	1 / 3
Distinctiveness Conflict Risk	The term 'reviewer' is somewhat specific and unlikely to broadly conflict, but 'Code review and quality assurance specialist' lives only in a secondary body block, not the description, so it could still overlap with general QA/review skills.	2 / 3
	Total	7 / 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: ruvnet/claude-flow
Path: .agents/skills/agent-reviewer/SKILL.md
Commit: 26c35b5

Reviewed: about 8 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.