CtrlK
BlogDocsLog inGet started
Tessl Logo

huashu-proofreading

三遍审校降低AI检测率,让文章更有人味。当用户提到"AI味太重"、"像AI写的"、"降低AI检测率"、"审校"、"自然一些"时使用。

69

Quality

83%

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

77%Weight 40%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is strongly actionable with a clear three-pass workflow and concrete checklists/examples. It loses points on conciseness from duplicated examples and a marketing footer, and on progressive disclosure because it is a long monolithic document with no bundle-file offloading.

Suggestions

Remove the duplicated before/after examples: keep them only in the 6-category taxonomy and drop the redundant 常见错误与改写示例 section (or vice versa), and delete the non-task marketing footer to reclaim tokens.

Split the detailed 素材库 usage and full example banks into reference bundle files under references/, keeping SKILL.md a lean overview that links to them one level deep.

Tighten the 审校工具 and 成功案例 sections to essential guidance, or move them to a reference file, to improve token efficiency.

DimensionReasoningScore

Conciseness

Mostly useful domain guidance, but ~300 lines with duplicated before/after examples (the same Claude Code rewrites appear in both the 6-category taxonomy and 常见错误与改写示例) and a non-task marketing footer ('30万+粉丝', '代表作…') that could be trimmed; not a 1 because it avoids explaining basic concepts Claude already knows.

2 / 3

Actionability

Highly actionable for an instruction skill: per-pass checkbox checklists, a concrete 6-category AI-flavor taxonomy with ❌/✅ rewrites, specific rewrite heuristics ('超过30字的拆成2-3句'), and a copy-paste-ready Grep invocation (pattern/path/output_mode/-n) for the 素材库.

3 / 3

Workflow Clarity

A clearly sequenced three-pass process (内容审校 → 风格审校 → 细节打磨) plus a 最终检查 validation read-through, each with its own checklist and method; the final read-through-and-rewrite loop serves as the validation checkpoint.

3 / 3

Progressive Disclosure

Well-organized with headers and one-level external references in 参考资源, but it is a large monolithic SKILL.md (~300 lines) with no bundle files and substantial content (detailed examples, 素材库 usage) kept inline that could be split out; above 1 due to clear structure, below 3 due to lack of one-level-deep bundle references and inline redundancy.

2 / 3

Total

10

/

12

Passed

Description

90%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A well-constructed description: third-person, concise, with an explicit when-trigger clause and natural user-facing trigger terms targeting a distinct niche. Its only weakness is that the 'what' names the method and goals rather than listing several concrete actions.

DimensionReasoningScore

Specificity

Names the domain and a concrete method ('三遍审校降低AI检测率,让文章更有人味') but does not enumerate multiple distinct concrete actions as a score-3 would; not vague, so above 1.

2 / 3

Completeness

States what it does (three-pass proofreading to lower AI detection / add human feel) AND when to use it via the explicit '当用户提到…时使用' clause with listed triggers, matching the what+when anchor.

3 / 3

Trigger Term Quality

Explicit quoted natural phrases a user would actually say ('AI味太重', '像AI写的', '降低AI检测率', '审校', '自然一些') give good coverage of real trigger terms.

3 / 3

Distinctiveness Conflict Risk

The AI-detection/humanizing niche with triggers like '降低AI检测率' is clearly distinct and unlikely to fire for general writing or editing skills.

3 / 3

Total

11

/

12

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
alchaincyf/huashu-skills
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.