A well-executed, instruction-only skill body: a clearly sequenced commit/push workflow with validation gates and failure handling, concrete commands, and correct deferral of message-format rules to a sibling skill. The only weaknesses are mild cross-section redundancy and the absence of an explicit commit-command example.

Suggestions

Deduplicate the fork-remote and issue-title-shortcut rules, which currently appear in both 快速指导 and 高信号规则/关键陷阱 — state each rule once and cross-reference it to trim tokens.

Add the concrete commit invocation (e.g. git commit -m "<message>") alongside the existing push commands so the execute path is fully copy-paste ready.

Merge 关键陷阱 items that merely invert 高信号规则 rules, keeping only the non-obvious pitfalls (e.g. hook-modified files, no-upstream push failure).

Dimension	Reasoning	Score
Conciseness	The body assumes Git competence with zero concept explanations and every section carries operational content, but fork-remote identification and the issue-title shortcut rules are each restated across 快速指导 and 高信号规则/关键陷阱, which could be trimmed.	4 / 5
Actionability	Gives concrete commands ('git branch --show-current', 'git remote get-url --push <remote>', 'git push -u <fork-remote> HEAD') and a worked commit-message example, but never shows the actual commit invocation (e.g. git commit -m "…"), leaving a minor gap versus copy-paste-ready guidance.	4 / 5
Workflow Clarity	The 12-step sequence has explicit pre-flight checks (branch/status, remote-role identification), a user-confirmation gate before side-effecting operations, error-recovery handling ('远端领先、分叉或被拒绝：停止默认推送，先向用户说明'), and a post-push report step, with 关键陷阱 serving as a failure-mode checklist.	5 / 5
Progressive Disclosure	Well-organized sections with no bundle files needed; the body appropriately defers commit-message format rules to git-commit-specification instead of duplicating them, and the 延伸阅读 cross-references are one level deep and clearly signaled.	5 / 5
	Total	18 / 20 Passed

Description

91%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 description: concrete third-person actions, an explicit use-when clause with quoted natural trigger phrases, and a clear niche in the BK-CI commit/push workflow. Only minor gaps in action coverage and slight trigger overlap with related git skills keep it from full marks.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions in third person — '根据当前变更生成一句简短提交内容、先展示给用户确认、确认后再 git commit，并默认推送到个人 fork 远端' — but omits parts of the workflow (remote-role identification, safety checks), so it falls just short of the comprehensive 5 anchor.	4 / 5
Completeness	Explicitly answers both 'what' (generate short message from diff, confirm with user, then commit, push to fork by default) and 'when' ('当用户要求…时优先使用') with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	Quotes natural user phrasings verbatim — '帮我 commit / push / 提交并推送代码 / 发到远端' — covering English command words and multiple Chinese synonyms users would actually say, matching the comprehensive-synonyms anchor.	5 / 5
Distinctiveness Conflict Risk	The confirm-then-commit, fork-default-push niche is clearly distinct, but the 'commit' trigger overlaps a sibling commit-message-writing skill, leaving minor conflict risk rather than the minimal risk of the 5 anchor.	4 / 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: TencentBlueKing/bk-ci
Path: ai/skills/git-commit-and-push/SKILL.md
Commit: d5420f2

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.