pr-review

仅当用户明确手动指定使用 pr-review skill 时触发；不要因为用户传入 PR 链接、要求 review 或要求代码审查而自动触发。

Quality

77%

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

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/system/pr-review/SKILL.md

Quality

Content

92%Weight 40%Scale 1-3

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

The content is an actionable, well-sequenced review workflow with strong conciseness and an explicit validation gate. Its main weakness is progressive disclosure: many referenced detail files are absent from the bundle.

Suggestions

Provide the referenced bundle files (backend-quality/*, frontend-quality/*, style/*) under references/ so the checklist links resolve, or inline the essential rules and drop dead links.

Consider moving the large i18n static-key rule and the full GitHub API review payload into reference files to keep SKILL.md a lean overview.

Verify cross-tree paths like ../../design/api/index.md are resolvable from the skill's installed location.

Dimension	Reasoning	Score
Conciseness	The body is lean and instructional — staged sections, command snippets, checklists, and a template — assuming Claude's competence without explaining basics; even the densest part (the i18n static-key rule) is skill-specific domain knowledge Claude would not already know.	3 / 3
Actionability	It provides fully executable, copy-paste-ready guidance throughout: concrete gh/git worktree commands, a real GitHub API review POST payload, a command reference table, and a report template.	3 / 3
Workflow Clarity	A clearly sequenced multi-stage process (Step 0 → Stage 1 → parallel Stages 2–6 → report) with an explicit validation gate (⚠️ 阶段门控) that aborts to reporting on failure and an error-recovery path asking the author to fix.	3 / 3
Progressive Disclosure	References to ~11 detail files (e.g. ./backend-quality/security.md, ./style/front.md) are clearly signaled one level deep, but none of those files exist in the skill bundle (references/, scripts/, assets/ absent and all paths verified missing), so navigation is broken; the body also inlines content that could be externalized.	2 / 3
	Total	11 / 12 Passed

Description

62%Weight 40%Scale 1-3

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 well-engineered as a trigger constraint and distinctiveness guard, but it omits any statement of what the skill actually does. Adding a capability clause would lift completeness and specificity.

Suggestions

Lead with a concise capability statement, e.g. 'Systematically reviews Pull Requests across requirements, backend/frontend quality, style, tests, and regression risk.', before the trigger constraint.

Keep the explicit 'only when manually named as pr-review' trigger clause — it is the description's strongest aspect.

Ensure the description answers both 'what' and 'when' in a single sentence to reach completeness level 3.

Dimension	Reasoning	Score
Specificity	The description only states a meta trigger rule ("手动指定使用 pr-review skill 时触发") and never names any concrete capability of the skill such as reviewing PRs, analyzing diffs, or checking code quality, matching the vague/no-actions anchor.	1 / 3
Completeness	The "when" is explicit and detailed, but the "what" (the skill performs PR code review) is never stated, so it cannot reach 3; it is above 1 because the trigger guidance is substantial rather than missing.	2 / 3
Trigger Term Quality	It enumerates natural trigger phrasings users actually say — "pr-review", "$pr-review", "使用 pr-review skill", "PR 链接", "review", "代码审查" — giving good coverage of common variations.	3 / 3
Distinctiveness Conflict Risk	The explicit anti-trigger framing (do NOT auto-trigger on PR links / review / code-review requests) carves a clear niche and makes false-trigger conflict risk low.	3 / 3
	Total	9 / 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
relative_links	Relative link issues: 11 missing, 2 suspicious	Warning

	Total	15 / 16 Passed

Repository: labring/FastGPT
Path: .agents/skills/system/pr-review/SKILL.md
Commit: dc97c08

Reviewed: 9 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.