CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

repository-module-architecture

处理 BK-CI 代码库接入、代码库授权、Webhook、SCM 集成、PAC 开关与仓库类型差异时使用。当用户提到 Git/SVN/GitHub/TGit、Webhook、仓库认证、PAC 代码库或提交记录时优先使用。

Quality

82%

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

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 well-organized, admirably lean router-style overview that classifies problems, states high-signal domain rules, and flags pitfalls — but its two reference pointers are dangling (the referenced files are absent from the bundle), and the body itself carries no concrete specifics (services, APIs, table names) to fall back on. Fixing or creating the reference files is the single highest-impact improvement.

Suggestions

Create the referenced files `reference/1-repository-foundation.md` and `reference/2-webhook-pac-auth.md` (or correct the paths to wherever they actually live, e.g. a `references/` directory) so the routing pointers resolve — currently the skill's core navigation dead-ends.

Add a minimal concrete mapping in the body (repository type → owning service/implementation, and the names of the key detail tables and auth methods) so the skill remains actionable even if a reference file is unavailable.

Ground the 高信号规则 section with one or two specific examples (e.g. a named API or field where the repo types diverge) to convert directional rules into executable guidance.

DimensionReasoningScore

Conciseness

The body is lean and signal-dense — short bullet sections (适用场景, 不适用场景, 高信号规则, 关键陷阱) with zero padding and no explanation of concepts Claude already knows, e.g. "PAC 是仓库侧开关和上下文来源,不等于 YAML 转换实现本身" states only non-obvious domain facts. Every token earns its place, matching the top anchor.

5 / 5

Actionability

The routing guidance is concrete in shape — "仓库基础与类型差异:看 `reference/1-repository-foundation.md`" — but the guidance never reaches executable specifics: no service names, API endpoints, table names, or field-level details, and the two referenced files do not exist in the bundle (no `reference/` or `references/` directory), so the actionable path dead-ends. This fits 'some concrete guidance but incomplete; missing key details'; it is above 2 because the classification and cross-check directives ("只要改仓库类型...就要同时看仓库明细表、认证方式和 Webhook 行为") are genuinely directive.

3 / 5

Workflow Clarity

快速指导 lays out a clear sequence — classify the problem into one of two mainlines, read the matching reference, apply the cross-check rule when changing repo type/auth/PAC, and hand off to `yaml-pipeline-transfer` when the problem enters YAML parsing — with the cross-check step acting as a checkpoint. Fits 'clear sequence with most checkpoints present'; not 5 because there are no explicit validation or error-recovery steps, though no destructive/batch operations exist that would cap the score at 3.

4 / 5

Progressive Disclosure

The body itself is well structured as an overview with clearly signaled, one-level-deep pointers ("仓库基础:`reference/1-repository-foundation.md`", "Webhook 与 PAC:`reference/2-webhook-pac-auth.md`"), which would merit 4 — but neither file exists in the bundle (there is no `reference/` or `references/` directory at all), so navigation from the overview leads nowhere. Dangling references are more than a 'minor organization gap', landing this at 'some structure but could be better organized'; it stays above 2 because the overview/pointer architecture itself is correct and nothing that belongs in separate files is inlined.

3 / 5

Total

15

/

20

Passed

Description

83%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 a strong two-part statement: a concrete enumeration of what the skill handles in the BK-CI repository domain, followed by an explicit 'use when the user mentions...' trigger clause with good platform-synonym coverage. Minor room to improve: add a couple more natural phrasings (clone, merge request, credentials) and sharpen the 'what' with action verbs.

DimensionReasoningScore

Specificity

Enumerates several concrete capability areas — "代码库接入、代码库授权、Webhook、SCM 集成、PAC 开关与仓库类型差异" (repository onboarding, authorization, Webhook, SCM integration, PAC switch, repo type differences) — which matches the anchor 'lists several specific actions; minor gaps in coverage'. It falls short of 5 because the coverage is stated as domain nouns rather than concrete actions (no verbs like configure/troubleshoot/audit), and it is not a fully comprehensive list; it is above 3 because far more than 1-2 concrete areas are named.

4 / 5

Completeness

It explicitly answers both: what — "处理 BK-CI 代码库接入...PAC 开关与仓库类型差异" — and when — "当用户提到 Git/SVN/GitHub/TGit、Webhook、仓库认证、PAC 代码库或提交记录时优先使用", a clear 'Use when...' clause with concrete trigger phrases, exactly matching the top anchor. A missing or only implicit 'when' clause would cap it at 3, but here it is fully explicit.

5 / 5

Trigger Term Quality

The trigger clause "当用户提到 Git/SVN/GitHub/TGit、Webhook、仓库认证、PAC 代码库或提交记录时" gives good natural keyword coverage including platform synonyms (Git, SVN, GitHub, TGit), which fits 'good keyword coverage; a few natural terms missing'. Not 5 because common user phrasings like clone, merge request, 代码仓库, or repo credential variants are absent.

4 / 5

Distinctiveness Conflict Risk

Anchored to a clear niche (BK-CI repository module, PAC, TGit) with distinct triggers, fitting 'mostly distinct; minor overlap risk with closely related skills'. Not 5 because triggering on a bare mention of "Git" could pull in ordinary Git-usage questions that the body explicitly excludes, creating minor overlap with generic Git skills.

4 / 5

Total

17

/

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

Path

ai/skills/repository-module-architecture/SKILL.md

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.