wechat-article

AI Berkshire skill: 微信公众号文章：作者-编辑-读者三Agent协作. Source: skills/wechat-article.md.

Quality

43%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./codex-skills/wechat-article/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.

This is a well-structured, actionable instruction skill with concrete prompt templates, commands, and revision rules; its main weaknesses are inline-heavy content that would benefit from referenced bundle files and the lack of an explicit validation feedback loop for the batch image-extraction step.

Suggestions

Move the three full agent prompt templates (Author/Editor/Reader) into separate files under references/ (e.g. prompts/author.md) and link to them from the body, keeping only their purpose and invocation notes inline.

Add an explicit validation/retry checkpoint to the image-extraction workflow (e.g. verify each rendered PNG is ≥500KB and re-render at higher DPI if not) so the batch operation has a feedback loop.

Deduplicate the image-extraction instructions, which currently appear in both 阶段二 (配图要求) and 阶段四 (第三步), into a single authoritative section to tighten token use.

Dimension	Reasoning	Score
Conciseness	The body is mostly task-focused and avoids explaining basic concepts, but the full prompt templates (Author/Editor/Reader) and the duplicated image-extraction instructions (in 阶段二 and again in 阶段四) add padding that could be tightened or factored out.	3 / 5
Actionability	It gives concrete, executable guidance throughout: specific pdftoppm/PIL commands with DPI values, LaTeX formatting rules, file-naming patterns, and ready-to-use agent prompt templates with explicit input slots.	4 / 5
Workflow Clarity	The four-phase process (research → draft → review → finalize) is clearly sequenced with sub-steps, a feedback-integration priority table, and revision rules; the main gap is the absence of an explicit validate/retry checkpoint for the batch image-extraction operation.	4 / 5
Progressive Disclosure	No bundle files (references/scripts/assets) exist, yet the body inlines three large prompt templates and a multi-table spec that would read better as separate referenced files; structure is present via clear section headers, but content that should be split is inline.	3 / 5
	Total	14 / 20 Passed

Description

23%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 a bare, metadata-laden label rather than a capability statement: it names the niche and a source file but omits concrete actions and any 'Use when...' trigger. It would rarely surface at the right moment because a user request phrased in natural terms would not match it well.

Suggestions

Rewrite as a third-person capability statement listing concrete actions, e.g. 'Drafts, edits, and reader-reviews publish-ready WeChat Official Account articles via an author-editor-reader three-agent workflow.'

Add an explicit trigger clause such as 'Use when the user asks to write, draft, or polish a WeChat Official Account (公众号) article, or wants a deep-research explainer formatted for 公众号.'

Remove the 'AI Berkshire skill:' prefix and the 'Source: skills/wechat-article.md.' provenance note from the description; these are metadata, not capability or trigger information.

Dimension	Reasoning	Score
Specificity	The description ('AI Berkshire skill: 微信公众号文章：作者-编辑-读者三Agent协作. Source: skills/wechat-article.md.') only names the domain and the three-agent collaboration theme; no concrete actions or capabilities are listed, leaving the 'what' implicit and abstract.	1 / 5
Completeness	It offers only a vague 'what' (a three-agent WeChat article collaboration) and no 'when' / 'Use when...' clause, so per the rubric guidance a missing trigger clause caps completeness; a 3-anchor 'clear what' is not reached because the what itself is terse and unaugmented by actions.	2 / 5
Trigger Term Quality	It includes the natural term '微信公众号文章' which a user might say, but adds the metacommentary 'AI Berkshire skill' and a source-path note rather than expanding the natural trigger phrases users would actually use.	2 / 5
Distinctiveness Conflict Risk	The WeChat-article niche is fairly specific and unlikely to collide with unrelated skills, but the generic 'AI Berkshire skill' prefix and lack of explicit triggers leave moderate overlap risk with other writing/research skills.	3 / 5
	Total	8 / 20 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, 1 suspicious	Warning

	Total	15 / 16 Passed

Repository: xbtlin/ai-berkshire
Path: codex-skills/wechat-article/SKILL.md
Commit: be46832

Reviewed: 3 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.