wx-mp-hunter

微信公众号内容抓取。如下三个场景必用本技能：(1) 用户提供 mp.weixin.qq.com 文章链接 -> 直接用本技能 fetch 全文（不要用浏览器或 web_fetch）；(2) 用户提供公众号名称、要求获取该账号过去数小时的发布列表 -> 用本技能 posts-list；(3) 用户提供 mp.weixin.qq.com/mp/homepage 专题页/主题页链接或 mp.weixin.qq.com/mp/appmsgalbum 合集链接，要求采集该页面全部文章 -> 用本技能 homepage 采集目录链接

Quality

85%

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

Quality

Content

82%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 skill body is highly actionable with executable commands, option tables, and output schemas for all three workflows, plus real validation checkpoints for batch operations. Its main weaknesses are mild verbosity in restating the no-login rationale and not explicitly signaling the bundled script file for navigation.

Suggestions

Add a one-line pointer near the top (e.g., '实现见 scripts/wx_mp_hunter.py') so the bundle file is explicitly signaled rather than implicit.

Condense the three repeated '不需要登录态' rationale blocks into a single short note, keeping only the per-command behavioral differences.

Consolidate the batch-fetch guidance into one explicit validate→batch→stop-on-3-failures checklist that applies across both posts-list and homepage workflows.

Dimension	Reasoning	Score
Conciseness	The body is largely efficient—command signatures, option tables, and JSON output examples earn their tokens—but it re-explains some mechanics Claude can infer (e.g., repeated '不需要登录态' rationale, wind-down of the 302/nwr_flag redirect chain) and could be tightened.	4 / 5
Actionability	It provides copy-paste-ready command signatures (wx-mp-hunter fetch <url> [--html] [--download-images] [--output-dir <dir>]), full option tables, and concrete JSON output schemas for all three sub-commands, covering the common cases completely.	5 / 5
Workflow Clarity	The homepage and 全文采集 sections give explicit numbered sequences with validation checkpoints (小样本验证, 连续失败3篇停止批量, 验证码不绕过), and the Agent 行为约束 section enforces wait-for-result discipline; however the overall page lacks one consolidated end-to-end checklist and the homepage batch loop's retry feedback loop is implicit rather than fully spelled out.	4 / 5
Progressive Disclosure	Structure is well organized with clear section headers per sub-command and the executable implementation correctly lives in scripts/wx_mp_hunter.py while SKILL.md stays an overview, but the body does not explicitly link to or signal the script file, so navigation to the bundle is implicit.	4 / 5
	Total	17 / 20 Passed

Description

87%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 strong: it states the domain, enumerates three concrete trigger scenarios mapped to specific sub-commands, and scopes out non-supported cases. It is slightly verbose and uses scenario framing rather than a crisp capability list, but trigger quality and distinctiveness are excellent.

Suggestions

Lead with a concise third-person capability statement (e.g., '抓取微信公众号文章全文、账号发布列表与专题页目录链接') before the numbered scenarios so the 'what' reads as a capability list rather than only scenarios.

Add common natural synonyms like '公众号文章' or '微信文章采集' alongside the URL-based triggers to broaden keyword coverage.

Dimension	Reasoning	Score
Specificity	The description names the domain (微信公众号内容抓取) and lists several concrete actions across three scenarios—fetch full text by URL, list posts of an account over N hours, and collect article links from homepage/album pages—but the actions are framed as user-input scenarios rather than enumerated capabilities, leaving minor coverage gaps.	4 / 5
Completeness	It explicitly answers both what (抓取微信公众号内容) and when via three numbered '必用本技能' trigger scenarios mapping user inputs to sub-commands (fetch / posts-list / homepage), satisfying the 'Use when...' bar with concrete trigger phrases.	5 / 5
Trigger Term Quality	It surfaces natural triggers users would actually say ('mp.weixin.qq.com 文章链接', 公众号名称, mp.weixin.qq.com/mp/homepage 专题页/主题页链接, mp/appmsgalbum 合集链接) with good keyword coverage, though it is missing common synonym phrasings like '公众号文章' or '微信文章'.	4 / 5
Distinctiveness Conflict Risk	It carves a clear niche bound to mp.weixin.qq.com URLs and WeChat Official Account content, with explicit non-support boundaries (视频号/comments/engagement metrics) that minimize overlap with other skills.	5 / 5
	Total	18 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
relative_links	Relative link issues: 1 missing	Warning

	Total	13 / 16 Passed

Repository: TeamWiseFlow/xiaobei
Path: crews/main/skills/wx-mp-hunter/SKILL.md
Commit: fdc03d6

Reviewed: 11 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.