CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

tapd-story-commit

迭代执行流水线提交阶段。在主会话中直接执行(不走 subagent),负责: 变更统计、成本数据汇总、构建 commit 信息、git 提交、TAPD 状态更新。 是 pipeline 中唯一在主会话执行的子 skill。

Quality

77%

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 ./.agents/skills/tapd-story-pipeline/tapd-story-commit/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

81%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.

The body is a strong, actionable workflow document: clearly sequenced steps with explicit validation gates, idempotency guarantees, and error-recovery handling, plus concrete commands and output templates. Remaining improvements are minor — removing the duplicated phase-update step, fixing the 3.1/3.2 subsection numbering, and giving an explicit git add command for step 7's default precise-staging path.

DimensionReasoningScore

Conciseness

The body is dense and table-driven with no explanations of concepts Claude already knows, but '步骤 4 更新状态' and '步骤 9 更新 meta.yaml.phase 为 committed' duplicate each other, and section 2 contains misnumbered subsections '3.1/3.2' — minor instances that could be trimmed, so anchor 4 rather than 5.

4 / 5

Actionability

Concrete commands and formats throughout ('git diff --numstat <baseline>', the stories_update parameter table with v_status="for test", '[cost-reconcile] jsonl_events=<N>...', and a full commit.md template). The gap is step 7's default path ('精确暂存 Code scope + WORKDIR 产物'), which specifies no exact git add command — mostly executable with minor gaps, matching anchor 4.

4 / 5

Workflow Clarity

A fixed explicit ordering ('1.5 对账 → 2 成本汇总 → 3 更新 stats → 6 写 commit.md → 7 git commit'), explicit validation gates with fix-and-retry feedback loops (cost reconciliation by ts_event dedup; the commit.md gate '任一缺失或为空则补写完整后再继续'), non-blocking error handling for TAPD failure, and a full idempotency/re-entrancy table — matching the top anchor.

5 / 5

Progressive Disclosure

References are well-signaled one level deep via a dedicated table with purpose and '何时读取' columns, and inline content (stats table, commit.md template) is appropriately placed in SKILL.md. However, the referenced files ('../references/commit-conventions.md', '../references/context-and-meta-template.md') use parent-relative paths and no references/ directory exists in the bundle, leaving them unverifiable — a minor organization gap that keeps this at anchor 4.

4 / 5

Total

17

/

20

Passed

Description

63%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 and comprehensive about what the skill does, with a clearly defined niche in the TAPD iteration pipeline. Its main weaknesses are the absence of any 'Use when' trigger guidance and heavy reliance on orchestration-internal jargon rather than natural user-facing trigger language.

Suggestions

Add an explicit 'Use when' clause stating when this skill should be invoked (e.g., '当需求 phase 为 validated 且需要提交代码并更新 TAPD 状态时使用') to raise completeness above 3.

Include natural user-facing trigger phrases such as '提交代码', '生成 commit message', '更新 TAPD 需求状态' alongside the internal pipeline terminology.

Reduce the orchestration-positioning text ('不走 subagent', '唯一在主会话执行的子 skill') or move it into the body, since it consumes description budget without aiding triggering.

DimensionReasoningScore

Specificity

The description lists five concrete actions — '变更统计、成本数据汇总、构建 commit 信息、git 提交、TAPD 状态更新' — which comprehensively covers the stage's capabilities, matching the top anchor rather than leaving minor gaps.

5 / 5

Completeness

The 'what' is clear (the five enumerated duties), but there is no 'Use when' clause or equivalent trigger guidance — only a weak positional implication ('是 pipeline 中唯一在主会话执行的子 skill'). Per the rubric guideline, a missing 'Use when' caps completeness at 3.

3 / 5

Trigger Term Quality

Core terms appear in both languages ('提交/commit', 'git 提交', 'TAPD', '流水线/pipeline'), but the phrasing is dominated by orchestration-internal jargon ('主会话', '不走 subagent', '唯一在主会话执行的子 skill') that a user would not naturally say, and common variations like '提交代码' are missing. It sits between anchors 3 and 4, closer to 3.

3 / 5

Distinctiveness Conflict Risk

The TAPD pipeline commit-stage niche is clear and mostly distinct, but '构建 commit 信息' carries minor overlap risk with generic commit-message skills, so anchor 4 fits better than 5.

4 / 5

Total

15

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

TencentBlueKing/bk-bcs

Path

.agents/skills/tapd-story-pipeline/tapd-story-commit/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.