review

Review changed code for correctness, security, code quality, and performance. Use when the user asks to review code changes, a PR, or specific files. Invoke with `/review`, `/review <pr-number>`, `/review <file-path>`, `/review <pr-number> --comment` to post inline comments on the PR, `/review --fix` to apply the findings to your working tree, or `/review <pr-number> --resume` to continue an interrupted review of that PR instead of starting over. Add `--effort low|medium|high` to trade depth for speed (defaults to high for PRs, medium for local changes).

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

70%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.

An exceptionally actionable and well-sequenced workflow with strong validation gates and clean on-demand references. Its dominant weakness is verbosity: huge inline paragraphs and rationale asides inflate the body far past what the runtime needs.

Suggestions

Move the Step 6 multi-round re-review apparatus (convergence posture, ledger rulings, fix-induced/code-age/bounded-unbounded rules) into a gated reference file for high-effort PR re-reviews, leaving the core Step 6 format inline.

Strip the inline 'measured; DESIGN.md — …' parentheticals from rule statements — the skill already designates DESIGN.md as the home for that narrative, so carrying it here duplicates it and burns context.

Break the longest single-paragraph blocks (e.g. the convergence posture and open-Critical re-check) into short bulleted rules so the orchestrator can scan them instead of reading a wall of prose.

Dimension	Reasoning	Score
Conciseness	At ~1085 lines / 308KB the body is noticeably verbose, with many multi-thousand-word single paragraphs and pervasive 'measured; DESIGN.md —' justification asides that are rationale the skill itself says belongs in DESIGN.md, not runtime context.	2 / 5
Actionability	Fully executable throughout — copy-paste-ready CLI commands with exact flags, paths, exit codes (6/7/3), and named JSON fields, covering the common cases concretely.	5 / 5
Workflow Clarity	A clearly sequenced 9-step workflow with explicit validation checkpoints (check-coverage exit 3, fail-closed gates, whiff relaunch, open-Critical re-check before Approve) and feedback loops (422 recovery, relaunch-once) around the destructive posting operation.	5 / 5
Progressive Disclosure	References are well-signaled and one level deep (posting.md/persistence.md/aone.md gated by the verdict), but large conditional apparatus — the multi-round convergence posture, ledger rulings, fix-induced and code-age rules — is inlined in Step 6 rather than split into a re-review reference, so content that should be separate is inline.	3 / 5
	Total	15 / 20 Passed

Description

92%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, third-person description that concretely states what the skill does and when to use it, with rich invocation detail. Minor synonym coverage gaps keep trigger-term quality just below full marks.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — review across correctness/security/quality/performance, post inline comments on a PR, apply findings to the working tree, and resume an interrupted review — giving comprehensive coverage of the skill's capabilities.	5 / 5
Completeness	Explicitly answers both what ('Review changed code for correctness, security, code quality, and performance') and when ('Use when the user asks to review code changes, a PR, or specific files') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Natural terms users would say are well covered ('review code changes', 'a PR', 'specific files', 'review'), but a few common synonyms (audit, critique, feedback) are absent, so it falls just short of comprehensive.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear code-review niche with distinct triggers and explicit invocation modes (--comment/--fix/--resume/--effort) that minimize overlap with neighboring skills.	5 / 5
	Total	19 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (1086 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: QwenLM/qwen-code
Path: packages/core/src/skills/bundled/review/SKILL.md
Commit: 054eabb

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.