CtrlK
BlogDocsLog inGet started
Tessl Logo

khazix-writer

数字生命卡兹克(Khazix)的公众号长文写作skill。当用户需要撰写公众号文章、写稿子、续写文章、根据素材产出长文时使用。触发词包括但不限于:写文章、写稿子、帮我写、续写、扩写、公众号文章、长文、出稿、按我的风格写。即使用户只是说"帮我把这个写成文章"或"用我的风格写一下",只要上下文涉及内容创作和公众号输出,都应该触发。也适用于用户丢过来一个PDF、brief、新闻链接、语音转文字或任何素材说"帮我写篇文章"的场景。不要用于短内容(小红书帖子、推特、朋友圈)或纯标题摘要生成(那个用wechat-title skill)。

71

Quality

89%

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 1-5

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

The body is highly actionable and its four-step workflow with a layered validation system is exemplary for a style-transfer skill. Its weaknesses are token efficiency — duplicated rule statements and padded concept explanations — and progressive disclosure, since large chunks of reference-grade material (word banks, checklist detail) are inlined in SKILL.md rather than split into the existing reference files.

Suggestions

State each hard rule (punctuation ban, banned words) exactly once in a rules section and have the L1/L2 checklist reference them by name ('as defined in 绝对禁区') instead of restating the full lists twice.

Move the 推荐口语化词组 word banks and the L3/L4 checklist detail into references/ (e.g. a style_vocabulary.md and a self_check.md), keeping only a summary and a pointer in SKILL.md.

Trim concept tutorials (契诃夫之枪, 英雄之旅, 升番) to the one sentence that names the principle plus how to apply it, trusting the reader's existing knowledge of these well-known frameworks.

DimensionReasoningScore

Conciseness

Mostly novel style data Claude cannot know (recommended colloquial phrases, banned words with replacements), so much of the length earns its place — but there is real duplication: the punctuation ban appears three times (绝对禁区, L1-2, L2-4) and the banned-word list twice (绝对禁区 and L1-1), and concepts like 契诃夫之枪 and 英雄之旅 are explained at tutorial length, fitting 'mostly efficient but includes some unnecessary explanation or could be tightened'.

3 / 5

Actionability

For an instruction-only skill the guidance is fully concrete: banned words each carry an exact replacement ("'说白了' → 换成'坦率的讲'"), recommended phrases are enumerated, five article archetypes each specify 写法重心, and openings give literal first sentences ("故事是这样的。"), so a writer can execute every rule directly.

5 / 5

Workflow Clarity

A clear four-step sequence (理解素材 → 角色边界 → 写作 → 自检) culminating in an L1–L4 self-check with explicit pass standards per layer ("以上四项扫描零命中"), per-item fix instructions, and a feedback loop of find → fix → re-check, plus a defined 质检报告 output format.

5 / 5

Progressive Disclosure

The two references are real files, clearly signaled in a closing 参考资料 section and one level deep — but the ~400-line body inlines substantial material that belongs in those reference files (long 口语化词组 lists, the full L1–L4 checklist detail, extended style-technique essays), matching 'some structure but content that should be separate is inline'.

3 / 5

Total

16

/

20

Passed

Description

95%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 states what the skill does, when to trigger it (with an unusually rich set of natural trigger phrases and example user utterances), and where it should not apply. The only minor gap is that the listed actions are variations of a single core capability rather than a breadth of distinct actions.

DimensionReasoningScore

Specificity

Names the domain and several concrete actions — "撰写公众号文章、写稿子、续写文章、根据素材产出长文" — but these are mostly variants of one core capability (long-form writing) rather than comprehensive coverage of distinct actions, matching 'lists several specific actions; minor gaps'.

4 / 5

Completeness

Explicitly answers both 'what' (公众号长文写作skill with write/draft/continue/expand actions) and 'when' ("当用户需要...时使用" plus concrete trigger phrases and scenarios), and adds negative boundaries — matching the top anchor.

5 / 5

Trigger Term Quality

Comprehensive natural triggers including synonyms and colloquial variants — "写文章、写稿子、帮我写、续写、扩写、公众号文章、长文、出稿、按我的风格写" — plus example user phrasings and input types ("PDF、brief、新闻链接、语音转文字"), matching the 'comprehensive coverage including synonyms' anchor.

5 / 5

Distinctiveness Conflict Risk

A clear niche (a specific person's official-account long-form style) with explicit exclusions — "不要用于短内容(小红书帖子、推特、朋友圈)或纯标题摘要生成(那个用wechat-title skill)" — showing awareness of sibling skills and minimal conflict risk.

5 / 5

Total

19

/

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
KKKKhazix/khazix-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.