CtrlK
BlogDocsLog inGet started
Tessl Logo

git-commit-specification

编写 BK-CI Git 提交信息和整理提交边界时使用,例如选择 commit type、撰写 commit message、判断是否拆分提交和准备 PR 前自检。当用户要提交代码而不是讨论实现细节时优先使用。

62

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./ai/skills/git-commit-specification/SKILL.md
SKILL.md
Quality
Evals
Security

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.

A lean, well-structured instruction skill whose guidance is concrete enough to act on, with a clear pre-commit self-check checkpoint. Its weaknesses are redundant repetition of the high-risk-operations caveat, an over-explained conventional commit-type list, and thin message-format examples for the common cases.

Suggestions

State the high-risk Git history caveat once (in 不适用场景) and remove the repetitions in 快速指导 item 6, 关键陷阱, and 延伸阅读.

Compress the commit-type list to the non-obvious cases or a one-line enumeration, since the conventional feat/fix/refactor/perf/test/docs/chore meanings are knowledge Claude already has.

Add 1-2 complete message examples covering the common cases (with scope and issue number, and a subject-plus-body form) to make the format guidance copy-paste ready.

DimensionReasoningScore

Conciseness

The body is short and mostly on-point, but it includes material Claude already knows — the full conventional type list ("feat:新增能力 / fix:修复问题 / refactor:重构但不改外部行为…") — and repeats the high-risk-history caveat four times ("不适用场景", "快速指导" item 6, "关键陷阱", "延伸阅读"). This matches anchor 3 (mostly efficient but includes some unnecessary explanation or could be tightened) rather than anchor 2, since the padding is localized rather than pervasive.

3 / 5

Actionability

For an instruction-only skill the guidance is largely executable: a concrete format example ("feat: 添加流水线模板功能 #1234"), an explicit one-purpose-per-commit split rule, and a concrete pre-commit self-check list (聚焦、敏感信息、必要验证). It sits at anchor 4 rather than 5 because the common cases are only partially covered — no body/subject-structure examples, no guidance on scope or issue-number placement beyond one example.

4 / 5

Workflow Clarity

"快速指导" lays out an implicit sequence (classify the change → write the message → split mixed changes → self-check) and item 5 provides an explicit pre-commit validation checklist, which fits anchor 4 (clear sequence with most checkpoints present). It is below anchor 5 because the ordering is interleaved with principles (items 1 and 6) rather than presented as an ordered procedure with feedback on failure, and above anchor 3 because a concrete checkpoint does exist.

4 / 5

Progressive Disclosure

This is a simple, single-purpose skill under 50 lines with no bundle files (no references/, scripts/, or assets/ exist, and the body makes no file references that would need verification). Per the rubric's simple-skill guidance, well-organized sections alone justify a 5: the body is cleanly split into 适用场景, 不适用场景, 快速指导, 高信号规则, 关键陷阱, 延伸阅读 with nothing that belongs in a separate file.

5 / 5

Total

16

/

20

Passed

Description

75%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 is specific, bilingual in its trigger vocabulary, and explicitly answers both what the skill does and when to use it. Its main weakness is that the 'when' clause is a single condition rather than a richer set of trigger phrases, and a few natural synonyms are absent.

Suggestions

Expand the 'when' clause with concrete trigger phrases users would actually say, e.g. "Use when the user asks to write a commit message, pick a commit type, split staged changes into multiple commits, or self-check before pushing/preparing a PR".

Add natural synonym keywords such as "git commit", "提交信息", and "commit 规范" to broaden trigger coverage in both Chinese and English.

DimensionReasoningScore

Specificity

The description lists several concrete actions — "选择 commit type", "撰写 commit message", "判断是否拆分提交", "准备 PR 前自检" — alongside the domain (BK-CI Git commits). It falls at anchor 4 rather than 5 because "整理提交边界" is stated somewhat abstractly and the coverage of the skill's rules (e.g. message format, issue-number conventions) is implied rather than enumerated; it is above anchor 3 since well more than 1-2 concrete actions are named.

4 / 5

Completeness

Both parts are explicit: the 'what' ("编写 BK-CI Git 提交信息和整理提交边界…") and the 'when' ("当用户要提交代码而不是讨论实现细节时优先使用"). The 'when' clause is present and usable but is a single discriminating condition rather than a set of concrete trigger phrases, matching anchor 4 ('when' could be more explicit or specific) rather than anchor 5.

4 / 5

Trigger Term Quality

It includes natural terms users would say in this domain: "commit type", "commit message", "拆分提交", "提交代码", "PR". Coverage is good but a few natural variations are missing (e.g. "提交信息", "git commit", "写提交说明"), placing it at anchor 4 rather than the comprehensive synonym coverage of anchor 5, while clearly above the generic keyword level of anchor 3.

4 / 5

Distinctiveness Conflict Risk

The niche (BK-CI commit message conventions and commit-boundary hygiene) is fairly distinct with dedicated triggers like "commit type" and "拆分提交". Minor overlap risk remains with generic git skills or PR-preparation skills, so it fits anchor 4 (mostly distinct, minor overlap risk) rather than anchor 5's minimal-conflict clear niche.

4 / 5

Total

16

/

20

Passed

Validation

100%

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
Reviewed

Table of Contents

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.