Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
wx-md-article
v1.0.0Automatically generate WeChat public account articles from Markdown and upload them to the draft folder with clean, professional formatting and color rules.
⭐ 1· 212·0 current·0 all-time
bySundy Yang@yangchao228
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 code: scripts convert Markdown to HTML and call WeChat APIs to upload drafts. However, instead of declaring required credentials or prompting the user to supply them, the package includes a config.json with an appid and appsecret embedded — that is unusual and should be justified (example config is expected, but these look like real values).
Instruction Scope
SKILL.md and the scripts confine themselves to converting the provided input file, building HTML, and calling api.weixin.qq.com endpoints. They do not read unrelated system files or call external endpoints beyond the WeChat API. One scope note: the script echoes part of the access_token to stdout which can leak secrets into logs.
Install Mechanism
There is no remote-install step or downloads. This is an instruction+script package; nothing is fetched from arbitrary URLs at install time. Risk surface is limited to the included shell scripts being run locally.
Credentials
The skill declares no required environment variables or primary credential, yet config.json contains an appid and appsecret (and a default thumb_media_id). Bundling active credentials in the package is disproportionate and dangerous: the script will use those credentials to act on behalf of that WeChat account rather than the user's account. The skill should instead prompt for or document replacing these with the user's own credentials.
Persistence & Privilege
The skill does not request persistent or system-wide privileges (always:false). It writes temporary files under /tmp while running and cleans them up; it does not modify other skills or system configs.
What to consider before installing
This skill does what it says (converts Markdown and uploads to WeChat) but the package includes a config.json with an appid and appsecret embedded. Do NOT use those credentials. Before installing or running: (1) Replace the appid/appsecret in config.json with your own WeChat credentials (or modify the scripts to read credentials from a secure location or env vars); (2) Rotate the embedded credentials if you control the referenced account, and avoid uploading sensitive content while using third-party credentials; (3) Be aware the script prints part of the access_token to stdout — avoid running it where logs are public; (4) Inspect and run the scripts in an isolated environment first; (5) If you don't own the embedded appid, do not rely on or trust that account (it could be abused to collect drafts under that third-party account). If the author intended the config values to be placeholders, they should be clearly marked as such; absence of that clarification is the main reason this package is suspicious.Like a lobster shell, security has layers — review code before you run it.
latestvk97eq3441nv71d4m43351h5mrn82m2v2
License
MIT-0
Free to use, modify, and redistribute. No attribution required.