Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
WeChat2PDF
v0.1.2将微信公众号文章完美转换为高清排版的无防盗链 PDF 和 Markdown 文件，方便永久归档与离线阅读。
⭐ 0· 219·0 current·0 all-time
by@xi0ng8
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
Name/description match the included code: run.py fetches WeChat article HTML, downloads images (handling lazy-load/data-src), embeds images (base64) in HTML, writes assets and markdown, and optionally uses Playwright to render PDF. The declared requirements (requests, bs4, markdownify, lxml, playwright) are appropriate for the stated task.
Instruction Scope
SKILL.md contains a high-priority routing directive: 'must unconditionally trigger' whenever the user's message contains an mp.weixin.qq.com link. This is broader than a normal usage note and can cause the agent to invoke the skill in contexts the user didn't intend. Otherwise, the runtime instructions (install Python libs, run run.py, return file paths or attachments) are narrowly scoped to the conversion task and do not request unrelated files or secrets.
Install Mechanism
No automated install spec is included (instruction-only install guidance). Dependencies are standard PyPI packages and Playwright, which is heavyweight but expected for headless Chromium PDF rendering. No downloads from unknown/personal servers or obscure URLs are present.
Credentials
The skill requests no environment variables, no credentials, and no config paths. Network access to download the target article and images is required and is consistent with the stated purpose.
Persistence & Privilege
The skill itself does not set always:true and does not request special system-wide privileges, but the SKILL.md's 'CRITICAL ROUTING' instruction attempts to force prioritized invocation whenever a specific domain appears. That instruction increases the skill's effective invocation scope beyond normal and could amplify impact if misused.
What to consider before installing
This skill appears to implement exactly what it promises (fetch a WeChat article, download images, build HTML/MD/PDF). The main red flag is the SKILL.md instruction that tells the agent to always and unconditionally trigger on any mp.weixin.qq.com link — that can lead to unwanted automatic invocations. Before installing: 1) decide whether you want the skill to auto-run on every WeChat link (it may be triggered in contexts where you don't expect it); 2) be aware it will download remote images and write files to disk (assets and PDFs); 3) Playwright/Chromium is heavy and may require additional downloads; 4) run the code in a sandbox or review and run it locally first if you have privacy or safety concerns; and 5) consider legal/copyright implications of archiving third-party articles. If you proceed, remove or relax the unconditional routing directive if you prefer explicit user consent each time.Like a lobster shell, security has layers — review code before you run it.
latestvk976kvkz2t6hehncez15c6kry582vy81
License
MIT-0
Free to use, modify, and redistribute. No attribution required.