wechat-theme-extractor — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited wechat-theme-extractor (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
A bulleted imperative like {match} tells the agent to never reveal, disclose, or mention something to the user. Used adversarially it can instruct the agent to hide its tool calls or lie about what it did — stripping the transparency a user relies on to trust the agent.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Extract layout and style signals from a WeChat Official Account article, then generate a reusable theme for markdown-wechat-converter. The workflow derives typography, colors, spacing, and block styles from the article, writes the generated theme into markdown-to-wechat.html, and supports previewing the result.
Use this skill when the user:
mp.weixin.qq.com article URLmarkdown-wechat-convertermarkdown-to-wechat.html and previewedDo not use this skill when:
markdown-wechat-converter or markdown-to-wechat.htmlscripts/extract.py to fetch the article HTML and extract the title and js_content..extracted_content.html and analyze style traits directly from the HTML.markdown-wechat-converter.markdown-to-wechat.html in the current workspace and inject the new theme without breaking existing theme definitions.markdown-to-wechat.html in the browser for preview.User provides URL
-> run extractor script
-> produce .extracted_content.html
-> analyze styles
-> generate theme config
-> write into markdown-to-wechat.html
-> verify write result
-> open previewpython3 scripts/extract.py "https://mp.weixin.qq.com/s/xxxxx"The Python script only:
js_content body fragment.extracted_content.htmlThe AI handles:
markdown-to-wechat.htmlWhen updating markdown-to-wechat.html, follow these rules:
const assignments before editing.Generate a theme object that matches the target file's existing structure. Use this as a minimal reference shape:
{
id: "wechat-clean-blue",
name: "WeChat Clean Blue",
styles: {
body: {
fontFamily: "\"PingFang SC\", \"Helvetica Neue\", sans-serif",
fontSize: "16px",
color: "#2b2b2b",
lineHeight: "1.75",
backgroundColor: "#ffffff"
},
h1: {
fontSize: "24px",
fontWeight: "700",
textAlign: "center",
color: "#1f3a5f"
},
h2: {
fontSize: "20px",
fontWeight: "700",
color: "#1f3a5f",
borderBottom: "2px solid #9ec1ff"
},
blockquote: {
color: "#4a5568",
backgroundColor: "#f7fbff",
borderLeft: "4px solid #7fb3ff",
padding: "12px 16px"
}
}
}When writing:
markdown-to-wechat.htmlscripts/.extracted_content.html: extracted article body HTML with title and source URL comments at the top.extracted_content.html exists and is not emptyjs_content were extractedmarkdown-to-wechat.html~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.