CtrlK
BlogDocsLog inGet started
Tessl Logo

writing-skills

创建或修改 Clowder AI skill / MCP tool description 的元技能(含质量标准、范本、发布)。 Use when: 写新 skill、修改现有 skill、写/改 MCP tool description、验证 skill 质量; 或者功能实现中产出了 SKILL.md / cat-cafe-skills/ 新目录 / manifest.yaml skill 条目。 Not for: 使用 skill(直接触发对应 skill)。 Output: 新/更新的 SKILL.md + manifest 条目 + symlinks。 GOTCHA: 软硬同重——skill/MCP 质量 = 代码质量;不要写模型已知的通用教程,先过价值门禁。

62

Quality

73%

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 ./cat-cafe-skills/writing-skills/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%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 well-structured and actionable, with a concrete publish workflow that includes validation gates. Its weaknesses are redundancy (the value-gate message is repeated across sections) and progressive-disclosure drift: a body length that exceeds the skill's own 150-line smell plus dead internal references.

Suggestions

Dedupe the value-gate material: keep one canonical 价值门禁/T0-0 table and have other sections (Common Mistakes, T0) cross-reference it instead of restating it, to tighten conciseness.

Honor the skill's own 150-line smell by moving the T0 deep-dive and 概念边界指南 detail into refs/ files, leaving the body an overview that points one level deep.

Resolve the two *(internal reference removed)* links in 深入学习 and T0-1 — either restore real files or remove the dead entries so navigation is not broken.

DimensionReasoningScore

Conciseness

Mostly efficient tables and checklists, but the value-gate philosophy is restated three times (价值门禁 section, T0-0, and Common Mistakes) and there is exhortatory framing ("铁律:软硬同重") that could be trimmed — fitting the 3 anchor of 'mostly efficient but could be tightened' rather than the lean 4.

3 / 5

Actionability

Concrete executable guidance throughout: a publish checklist with real commands (pnpm sync:skills, pnpm check:skills, pnpm check), a copy-ready SKILL.md structure template, a 范本 table, and the MCP four-element checklist; as an instruction-only meta-skill the absence of code is fine, with only minor gaps keeping it at 4.

4 / 5

Workflow Clarity

The 发布检查清单 is a clear 5-step sequence (源文件 → 同步 → 注册 → 验证 → commit) with an explicit validation gate ("pnpm check:skills 全绿") and a conditional pre-commit check ("commit 前还必须跑 pnpm check"), matching the 4 anchor of clear sequence with most checkpoints present.

4 / 5

Progressive Disclosure

Good section structure and a dedicated 深入学习 table that signals refs one level deep, but the body is ~200 lines (over its own stated 150-line smell, inlining material the skill's own rule says should be 下沉 refs) and two refs are marked *(internal reference removed)* (dead navigation), fitting the 3 anchor of 'some structure but content that should be separate is inline'.

3 / 5

Total

14

/

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 strong: it uses the required Use-when / Not-for / Output three-piece format, names concrete authoring actions, and carries distinct, natural trigger terms. Voice is third-person with no first/second-person penalty. Minor room remains in synonym/file-extension saturation and edge-case overlap fencing.

DimensionReasoningScore

Specificity

Names concrete actions — "创建或修改 Clowder AI skill / MCP tool description" plus "质量标准、范本、发布" and Use-when triggers (写新 skill、修改、写/改 MCP、验证质量) — giving several specific actions with only minor coverage gaps, matching the 4 anchor rather than the fully comprehensive 5.

4 / 5

Completeness

Explicitly answers both what ("创建或修改…skill / MCP tool description 的元技能") and when ("Use when: …") with concrete trigger phrases, plus a full Not-for/Output/GOTCHA three-piece set — clearly above the 4 anchor which has only a weaker when.

5 / 5

Trigger Term Quality

Triggers like "写新 skill", "修改 skill", "SKILL.md", "cat-cafe-skills/", "manifest.yaml skill", "MCP description", "tool description" are natural phrases a user in this ecosystem would say, with good synonym coverage; not quite the comprehensive file-extension/synonym saturation of the 5 anchor.

4 / 5

Distinctiveness Conflict Risk

A clear niche (authoring skill/MCP files, not using them) with a Not-for clause ("Not for: 使用 skill") that fences off the main confusion; minor residual overlap with code-quality skills like quality-gate/tdd keeps it at 4 rather than 5.

4 / 5

Total

17

/

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.

Validation15 / 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
zts212653/clowder-ai
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.