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。

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./huashu-proofreading/SKILL.md

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.

A detailed, actionable three-pass proofreading workflow with strong concrete examples and checklists. Its main weaknesses are redundant example repetition that bloats token cost and a monolithic structure with no bundle files to offload reference material into.

Suggestions

Remove the duplicate ❌/✅ rewrite examples in '常见错误与改写示例' (and trim '第二遍审校：具体方法') since the same Claude Code patterns already appear in the 6-category section — keep one canonical set of examples.

Move the '个人素材库使用' guide and the error-example gallery into files under references/ and link to them one level deep, turning SKILL.md into a lean overview and improving progressive disclosure.

Tighten the '审校工具（可选）' section: either drop the external-tool name-drops (GPTZero/Originality.AI) or give a concrete invocation/check, rather than leaving them as '仅供参考'.

Dimension	Reasoning	Score
Conciseness	Mostly efficient domain-specific guidance (the 6 AI-腔 categories are genuinely non-obvious), but the same ❌/✅ Claude Code rewrite patterns are repeated across the 6-category section, '具体方法', and '常见错误与改写示例', adding padding that could be trimmed.	3 / 5
Actionability	Concrete checklists, specific thresholds (15-25字, <30字, <30% AI检测率), and a copy-paste-ready Grep invocation with parameters make the guidance executable; minor gaps remain in the tool-detection section which only name-drops external tools.	4 / 5
Workflow Clarity	The three-pass flow (内容审校 → 风格审校 → 细节打磨 → 最终检查) is clearly sequenced with per-pass checklists and an iterative read-aloud/re-read feedback loop; validation is subjective rather than objective, leaving a minor checkpoint gap.	4 / 5
Progressive Disclosure	No bundle files exist (references/scripts/assets are empty) and the body is a ~300-line monolith; reference-like content (the error-example gallery, the material-library usage guide) is inlined, and the listed 参考资源 paths point to the user's external filesystem rather than navigable bundle files.	3 / 5
	Total	14 / 20 Passed

Description

78%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.

A strong description that clearly answers both what the skill does and when to use it, backed by natural quoted trigger phrases. It is slightly held back by action framing that leans on goals rather than enumerated concrete capabilities and by the broadness of the '审校' trigger.

Dimension	Reasoning	Score
Specificity	Names the domain and 1-2 concrete actions ('三遍审校降低AI检测率', '让文章更有人味') but these read as goals rather than a comprehensive list of specific capabilities, matching the 'names domain and 1-2 concrete actions' anchor.	3 / 5
Completeness	Explicitly states what it does ('三遍审校降低AI检测率，让文章更有人味') and when to use it ('当用户提到...时使用') with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	Quotes five natural user phrases ('AI味太重', '像AI写的', '降低AI检测率', '审校', '自然一些') with good synonym coverage; a few common variants (e.g. '去AI味', '改一下AI腔') are missing, so it sits just below comprehensive.	4 / 5
Distinctiveness Conflict Risk	The AI-detection-reduction niche is distinct with specific triggers, but the bare term '审校' (proofreading) is broad and could overlap with general editing skills, leaving minor conflict risk.	4 / 5
	Total	16 / 20 Passed

Validation

100%

Warnings & errors only

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: alchaincyf/huashu-skills
Path: huashu-proofreading/SKILL.md
Commit: 49a55ba

Reviewed: 9 days ago

Table of Contents

Discovery Implementation Validation

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.