requesting-code-review

完成任务、实现重要功能或合并前使用，用于验证工作成果是否符合要求

Quality

62%

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/requesting-code-review/SKILL.md

Quality

Content

71%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.

The content is well-organized, concise, and actionable with a clear sequenced workflow and feedback loop, but its central progressive-disclosure reference (the code-reviewer.md template) is not bundled, which weakens both navigation and the copy-paste executability of the dispatch step.

Suggestions

Bundle the referenced code-reviewer.md template under references/ (or inline its dispatch prompt) so the core action is fully executable and the one-level reference resolves to a real file.

Promote the '评估 → 可以继续' gate from the example into an explicit workflow checkpoint step so the validation loop is formalized rather than implied.

Tighten the rationalization table or move it to a reference file to push conciseness from efficient to lean.

Dimension	Reasoning	Score
Conciseness	The body is sectioned and assumes Claude's intelligence (it never explains what code review is), with most tokens earning their place; only minor stretches like the rationalization table edge past lean, fitting 'efficient; minor instances that could be trimmed' below the 5-anchor's every-token-earns-its-place bar.	4 / 5
Actionability	It gives executable git commands (BASE_SHA/HEAD_SHA), a concrete placeholder list, and a worked example, but the core dispatch step delegates to an external 'code-reviewer.md' template that is not bundled, leaving a minor gap that keeps it below the fully copy-paste 5-anchor.	4 / 5
Workflow Clarity	The three-step sequence (get SHA → dispatch reviewer → handle feedback) is clear with a tiered feedback loop (Critical/Important/Minor) and an error-recovery path ('如果审查者有误，用技术理由反驳'), but the 'verdict → continue' checkpoint lives only in the example rather than as an explicit step, so it does not reach the 5-anchor's formalized validation checkpoints.	4 / 5
Progressive Disclosure	The body is well-sectioned and signals a one-level-deep reference ('参见模板：requesting-code-review/code-reviewer.md'), but no bundle directory exists and the referenced code-reviewer.md is not present, so navigation to the detail material fails — fitting 'references present but not clearly realizable' rather than the 4-anchor's mostly-clear references.	3 / 5
	Total	15 / 20 Passed

Description

53%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.

The description explicitly covers both when to use the skill and its verification purpose, but the 'what' is generic and lacks concrete review actions or natural 'code review' trigger terms, leaving it broad enough to risk overlap with testing/QA skills.

Suggestions

Add concrete actions to the description, e.g. '派遣代码审查子代理检查 diff、找出 Critical/Important/Minor 问题' so the 'what' is specific rather than a generic 'verify'.

Include natural trigger keywords users would actually say, such as '代码审查' / 'review my code' / '审查 diff', to improve trigger-term quality and distinctiveness.

Sharpen the niche so it cannot be confused with testing/QA, e.g. explicitly frame it as 'requesting a code-review subagent' rather than 'verifying work meets requirements'.

Dimension	Reasoning	Score
Specificity	The description names the domain ('验证工作成果是否符合要求') but the only action stated is the generic 'verify', with no concrete review actions like checking a diff or finding bugs, matching the 'names domain but actions minimal/generic' anchor rather than the 3-anchor which requires 1-2 concrete actions.	2 / 5
Completeness	Both 'what' ('验证工作成果是否符合要求') and 'when' ('完成任务、实现重要功能或合并前使用') are explicitly present, but the 'what' is a purpose statement rather than concrete actions, so it does not reach the 5-anchor's requirement of concrete trigger phrases for both.	4 / 5
Trigger Term Quality	It lists several natural trigger contexts ('完成任务、实现重要功能或合并前') but is missing common variations users would say such as 'code review' / '审查代码' / 'check my work', so it sits at 'some relevant keywords but missing synonyms' rather than the 4-anchor's good coverage.	3 / 5
Distinctiveness Conflict Risk	Tying the trigger to post-task / pre-merge moments gives some specificity, but the described action ('验证工作成果') is broad and could overlap with testing or QA skills, fitting 'somewhat specific but could still overlap' rather than the mostly-distinct 4-anchor.	3 / 5
	Total	12 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: jnMetaCode/superpowers-zh
Path: skills/requesting-code-review/SKILL.md
Commit: c68725b

Reviewed: 4 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.