CtrlK
BlogDocsLog inGet started
Tessl Logo

bk-monitor-tapd-dev

根据 TAPD 单据分析开发需求并生成方案。当用户需要开发 TAPD 需求/缺陷/任务时使用。

55

Quality

69%

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 ./bcs-services/bcs-project-manager/.cursor/skills/bk-monitor-tapd-dev/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

50%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 sketches a sensible workflow (get ID, validate branch, fetch ticket, analyze, confirm, develop) with a few genuinely concrete commands, but it reads like a compressed outline: steps 2-6 are grouped and vague, the resources section is duplicated, and none of the cited reference files exist in the bundle. It needs cleanup and either executable MCP call examples or verified reference files.

Suggestions

Delete the duplicated '📦 可用资源' section and the trailing meta note — keep one reference list with clickable relative paths.

Expand 'Step 2-6' into individually numbered steps with at least one concrete MCP call example (parameters, expected response shape) or defer explicitly to mcp-calls.md with what to look up there.

Add an error-handling branch for the ticket fetch (e.g., MCP call fails, ticket not found, no comments) and ensure the four referenced files actually exist in references/ so the skill is self-contained.

Fix the step numbering ('Step 1', 'Step 2-3', 'Step 4-6') so the sequence is continuous and each step is separately defined.

DimensionReasoningScore

Conciseness

The body is mostly efficient workflow text, but the '📦 可用资源' section is duplicated verbatim (identical 4-reference list twice) and a leftover meta line ('> 根据 SKILL.md 中的 IF-THEN 规则判断是否需要加载') pads the end. This is more than the 'minor instances that could be trimmed' of the 4 anchor, so 3 fits best.

3 / 5

Actionability

There are concrete elements ('git branch --show-current', the '#(\d+)' pattern, 'id.substring(2, 10)', and MCP tool names stories_get/bugs_get/tasks_get), but steps 2-6 are high-level direction ('调用 MCP 获取单据详情和评论', '按 dev-plan-template.md 生成方案') with no executable call examples or parameters. That matches the 'some concrete guidance but incomplete' anchor rather than 4.

3 / 5

Workflow Clarity

A sequence exists with a blocking branch-validation checkpoint and a user-confirmation gate before development, but steps are compressed into 'Step 2-3' and 'Step 4-6' groupings that leave gaps, and there is no error-recovery feedback loop (e.g., what to do when the MCP call fails or returns empty). This is the 'steps listed but checkpoints missing or implicit' anchor.

3 / 5

Progressive Disclosure

References are placed in a dedicated, clearly signaled section, but that section is duplicated verbatim, uses skill:// URIs rather than plain relative links, and the referenced files (branch-workflow.md, dev-plan-template.md, mcp-calls.md, tapd-mcp-guide.md) are not present in the bundle, so navigation is unreliable. This is 'some structure but could be better organized', below the 'minor organization gaps' of 4.

3 / 5

Total

12

/

20

Passed

Description

73%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 solid two-sentence what/when pair in third person, tied to a niche tool with natural trigger terms for the three TAPD work-item types. Its main weaknesses are modest action coverage (only analyze + generate plan) and missing synonyms/variations in the trigger phrasing.

Suggestions

Enumerate a few more concrete actions in the description (e.g., fetch ticket details and comments via MCP, verify the working branch matches the ticket ID, draft the dev plan) to lift specificity.

Add trigger synonyms users might say, such as English terms (TAPD story/bug/task) or '写开发方案/领单据', to broaden trigger_term_quality.

DimensionReasoningScore

Specificity

Names the domain (TAPD 单据) and 1-2 concrete actions ('分析开发需求并生成方案' — analyze requirements, generate a plan), matching the anchor for domain plus 1-2 actions. It stays below 4 because it does not list several specific actions (e.g., fetching ticket details, verifying branches, drafting the plan) and above 2 because the actions are concrete rather than generic.

3 / 5

Completeness

Both 'what' ('根据 TAPD 单据分析开发需求并生成方案') and an explicit 'when' ('当用户需要开发 TAPD 需求/缺陷/任务时使用') are present. The 'when' could be more explicit and specific about trigger contexts (writing a dev plan, picking up an assigned ticket), so it falls short of the 5 anchor.

4 / 5

Trigger Term Quality

'当用户需要开发 TAPD 需求/缺陷/任务时使用' covers the three natural TAPD work-item types a user would actually name, giving good keyword coverage. It is not 5 because common variations and synonyms (e.g., English 'story/bug/task', 'TAPD ticket') are missing.

4 / 5

Distinctiveness Conflict Risk

TAPD is a specific niche tracker and the trigger terms (TAPD 需求/缺陷/任务) are distinct, giving a clear niche with minimal conflict risk against generic development skills. It clearly matches the 5 anchor rather than 4, which assumes overlap with closely related skills.

5 / 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-bcs
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.