An exceptionally tight, actionable instruction skill: concrete commands, a precise per-section writing rubric, and illustrative format examples with no padding. The only meaningful gaps are a missing failure/retry loop around publishing and a visualization catalog that could be split into a reference file.

Dimension	Reasoning	Score
Conciseness	The body is dense, prescriptive, and free of concept explanations or padding — every rule is repo-specific guidance Claude could not infer, and the example blocks are minimal illustrations of format rather than filler. Anti-drift check: it clearly matches the "lean and efficient" anchor, not the "minor instances of over-explanation" anchor.	5 / 5
Actionability	Gives fully executable commands (gh pr view/create/edit with --body-file, git diff main...HEAD, reading the template path), a concrete table schema with an example row, and unambiguous per-section rules. As an instruction-only skill, the absence of executable code is not penalized since the guidance itself is directly actionable.	5 / 5
Workflow Clarity	The 5-step workflow is clearly sequenced (read template, find/create PR, gather context, write sections, publish) with an explicit confirm step, but validation is thin — there is a single "确认更新成功" checkpoint and no feedback loop describing what to do if publishing or verification fails. Not 5 for the missing explicit validate/retry checkpoints; not 3 because the sequence itself is fully specified and the operations are non-destructive.	4 / 5
Progressive Disclosure	A single well-organized file with clear section headers and pointed references to repo resources (.github/pull_request_template.md, .agents/skills/review-pr/surfaces.md) and the gen-changesets skill. The skill exceeds 50 lines and the ~75-line visualization-outline catalog is arguably inline content that could live in a reference file, so it falls just short of the "appropriately split" anchor.	4 / 5
	Total	18 / 20 Passed

Description

87%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, repo-scoped description that explicitly states both what it does and when to use it, with concrete natural-language triggers. Keyword coverage and action enumeration have only minor gaps, keeping it just short of perfect on those dimensions.

Dimension	Reasoning	Score
Specificity	Names several concrete actions — creating or updating PRs, writing each PR-template section concisely for reviewers, and producing the behavior-change/affected-users table — but coverage is not fully comprehensive (it does not enumerate the individual template sections or the visualization-outline guidance).	4 / 5
Completeness	Explicitly answers both: what (how to write each PR-template section concisely, including the behavior-change table) and when ("在 kimi-code 仓库里创建或更新 PR 时使用" is an explicit use-when clause with a concrete trigger).	5 / 5
Trigger Term Quality	Contains natural phrases a user would say such as "创建或更新 PR" and "PR 模板", giving good keyword coverage, though a few natural variants ("pull request", "PR description", "open a PR") are missing.	4 / 5
Distinctiveness Conflict Risk	The description is scoped to a specific repo's PR template and its distinctive behavior-change table, occupying a clear niche with minimal risk of triggering for the wrong skill.	5 / 5
	Total	18 / 20 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: MoonshotAI/kimi-code
Path: .agents/skills/write-pr/SKILL.md
Commit: 21406fb

Reviewed: 2 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.