writing-great-skills

编写和编辑优秀 skills 的参考：让技能可预测的词汇和原则。

Quality

31%

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 ./skills/productivity/writing-great-skills/SKILL.md

Quality

Content

55%

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

The body is a well-organized, concept-rich reference that assumes Claude's competence and avoids basic-concept padding, with clearly ordered internal procedures. Its main weaknesses are recurring restatement of a few ideas (conciseness) and a referenced GLOSSARY.md bundle file that is not present to verify the disclosure structure.

Suggestions

Consolidate the recurring mentions of duplication and premature completion so each concept has a single source of truth, tightening the body toward the lean anchor.

Ship the referenced GLOSSARY.md (and confirm it exists in references/) so the one-level-deep progressive disclosure is verifiable and the score can reach the clear-overview anchor.

Convert a few principle statements into concrete, copy-paste-ready checks (e.g. an explicit no-op test checklist) to lift actionability toward fully executable guidance.

Dimension	Reasoning	Score
Conciseness	The body is dense and mostly earns its tokens through novel vocabulary (predictability, legwork, completion criterion, leading word) without re-explaining basics Claude knows, but several passages restate the same idea across sections (e.g. duplication/premature completion recur) and could be tightened, matching the 'mostly efficient but could be tightened' anchor rather than the fully lean level.	2 / 3
Actionability	It gives concrete, specific guidance in places (e.g. '设置 disable-model-invocation: true', '把 skill 的 leading word 放前面', the fast->tight and a loop you believe in->red collapse examples), but much of it is principle/rationale rather than directly executable steps, and it is explicitly a reference skill, so it sits at 'some concrete guidance but incomplete' rather than fully copy-paste-ready.	2 / 3
Workflow Clarity	As a reference-only skill it has no multi-step destructive workflow, but its internal procedures are clearly sequenced with explicit checkpoints: the Pruning section orders 'single source of truth -> relevance -> no-op test, delete the whole sentence', and Failure modes gives a defensive ordering ('先 sharpen completion criterion...只有...才通过拆分'). The rubric's simple-skills note allows a 3 for clear single-purpose organization without a multi-step workflow.	3 / 3
Progressive Disclosure	The body signals a one-level-deep reference via '[GLOSSARY.md]' context pointers and states the skill discloses full definitions there, which is good structure, but the referenced GLOSSARY.md file is not present in the bundle (references/, scripts/, assets/ are absent), so the disclosure cannot be verified against an actual bundle and the top-level still carries substantial inline definitional material.	2 / 3
	Total	9 / 12 Passed

Description

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 terse human-facing summary rather than a model-invocation description, consistent with its disable-model-invocation flag. It lacks concrete actions, natural trigger terms, and any explicit "when to use" clause, so it scores low on specificity, trigger quality, and completeness.

Suggestions

Add concrete actions a user associates with the skill (e.g. 'Write, edit, and prune a SKILL.md; split and refactor skills') so specificity reaches the multi-action anchor.

Include natural trigger phrases users would actually say (e.g. 'Use when writing or improving a skill, pruning a SKILL.md, or deciding when to split skills').

Add an explicit 'when to use it' clause to answer both what and when; note that since disable-model-invocation is set, a richer model-facing description is only warranted if invocation mode changes.

Dimension	Reasoning	Score
Specificity	The description ("编写和编辑优秀 skills 的参考：让技能可预测的词汇和原则") names no concrete actions — it is an abstract one-line summary with no enumerated capabilities, matching the vague/abstract anchor and the bad example "Helps with documents".	1 / 3
Completeness	It states a rough "what" (a reference for writing skills) but provides no "when"/trigger clause at all — there is no "Use when…" guidance, so per the rubric guideline a missing explicit trigger caps it at 2, and the missing-what-OR-when anchor maps to 1.	1 / 3
Trigger Term Quality	It contains no natural keywords a user would say when they need this skill; "可预测" and "词汇和原则" are abstract conceptual terms rather than trigger phrases users actually utter, fitting the "No natural keywords" anchor.	1 / 3
Distinctiveness Conflict Risk	The niche (writing/editing skills) is somewhat specific and unlikely to broadly conflict, but the wording is generic enough that it could overlap with other skill-authoring guidance; this matches the "somewhat specific but could still overlap" anchor and is above the fully-generic level.	2 / 3
	Total	5 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 1 missing	Warning

	Total	15 / 16 Passed

Repository: vinvcn/mattpocock-skills-zh-CN
Path: skills/productivity/writing-great-skills/SKILL.md
Commit: f2b2464

Reviewed: 1 day 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.