code-reviewer

Elite code review expert specializing in modern AI-powered code

Quality

30%

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-reviewer/SKILL.md

Quality

Content

20%

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

The body is a verbose, persona-driven capability catalog that restates widely-known concepts and provides almost no executable guidance. Its one progressive-disclosure reference points to a non-existent file, and the only workflow is an abstract 10-phase list.

Suggestions

Collapse the capability/knowledge/behavioral bullet walls into a short overview, moving detailed checklists into real reference files (e.g. references/security-checklist.md) that actually exist.

Replace abstract phases with concrete, runnable guidance: example review commands (e.g. `semgrep scan --config p/owasp-top-10 .`), example review-comment templates, and specific verification steps.

Add validation checkpoints to the review workflow (e.g. 'run static analysis, confirm findings against the diff, then triage by severity') and fix or remove the broken `resources/implementation-playbook.md` reference.

Dimension	Reasoning	Score
Conciseness	The ~170-line body is a wall of capability bullets restating concepts Claude already knows (OWASP Top 10, SOLID, N+1, cyclomatic complexity, Clean Code), matching the level-1 'verbose; explains concepts Claude knows; padded' anchor. It cannot be level 2 because almost every token is restatement rather than novel, actionable guidance.	1 / 3
Actionability	The body describes rather than instructs: no executable code, commands, or concrete tool invocations, only abstract capability lists and phases like 'Apply automated tools for initial analysis'. The single concrete reference (`resources/implementation-playbook.md`) points to a file that does not exist, so it does not reach level 2.	1 / 3
Workflow Clarity	A 10-step 'Response Approach' sequence is present, but the steps are abstract phases with no validation checkpoints or concrete commands. It is above level 1 because a real numbered sequence exists, but below level 3 due to absent validation/feedback loops and lack of specificity.	2 / 3
Progressive Disclosure	Section headers provide some organization, but the bulk of capability content is inline monolithic text that should live in separate reference files, and the one reference (`resources/implementation-playbook.md`) is broken (no `resources/` directory exists). It is not level 3 (references not well-signaled/valid) and not level 1 (headers give real structure).	2 / 3
	Total	6 / 12 Passed

Description

40%

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 identifies a clear domain (code review) but is written as a persona headline rather than a capability-and-trigger statement. It is padded with buzzwords and omits any 'Use when…' guidance.

Suggestions

Replace the persona headline with concrete actions, e.g. 'Reviews code for security vulnerabilities, performance issues, and maintainability; generates structured review comments.'

Add an explicit trigger clause: 'Use when the user asks to review code, a pull request, or a diff, or wants feedback on code quality, security, or performance.'

Drop evaluative fluff like 'Elite' and 'modern AI-powered' in favor of specific, triggerable terms.

Dimension	Reasoning	Score
Specificity	The description ("Elite code review expert specializing in modern AI-powered code") states a persona rather than concrete actions, and relies on buzzwords ("Elite", "modern AI-powered") that the guidelines say to penalize. It is not a level-2 anchor because it names no specific actions comparable to 'Processes PDF files and extracts content'.	1 / 3
Completeness	It conveys a weak 'what' (a code-review expert) but has no 'Use when…' clause or equivalent trigger guidance, which the guidelines cap at 2. It is not level 1 because some 'what' is stated, and not level 3 because 'when' is entirely missing.	2 / 3
Trigger Term Quality	"code review" is a natural keyword a user would actually say, but the description omits common variations such as 'review my code', 'PR review', or 'review pull requests'. It is above level 1 because a genuine natural trigger term is present, but below level 3 due to thin coverage.	2 / 3
Distinctiveness Conflict Risk	"code review" is a recognizable niche, but the description lacks distinct triggers and could overlap with general coding skills. It is not level 3 (no clear distinct triggers) and not level 1 (the niche is more specific than 'Helps with code and documents').	2 / 3
	Total	7 / 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: sickn33/antigravity-awesome-skills
Path: skills/code-reviewer/SKILL.md
Commit: a3068b7

Reviewed: 1 day 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.