xhs-fav-export
wc3-code.mjs 是高度混淆的未声明 LLM 调用工具,与 SKILL.md 声称的「纯确定性代码」不符,存在阴影功能嫌疑
Why this conclusion was reached
0/4 dimensions flaggedDeclared resources and inferred behavior are broadly aligned.
6 lower-risk artifacts were extracted and still need context.
There is no explicit malicious chain in the report.
Dependency information is incomplete, so supply-chain confidence stays limited.
What drove the risk score up
wc3-code.mjs 完全未在 SKILL.md 中提及,且代码经过高度混淆(7804行压缩为1行)
SKILL.md 声称「全程确定性代码」「不调用 LLM 子会话」,但存在独立 LLM 调用工具
Most important evidence
wc3-code.mjs 高度代码混淆且未声明
wc3-code.mjs 是一个独立的 Node.js 工具文件,包含 Base64 解码器混淆逻辑和 LLM API 调用功能(WC3_LLM_ENDPOINT 环境变量),但未在任何文档中声明。7804行代码被压缩为单行,存在阴影功能风险。
wc3-code.mjs:1 SKILL.md 与实际代码功能存在偏差
SKILL.md 声称「全程确定性代码(Extension Relay 操作),不依赖本地 pipeline 子会话服务」,但存在独立的 wc3-code.mjs(LLM 调用工具)且未声明
SKILL.md:1 Declared capability vs actual capability
skill.mjs:RELAY_URL 调用 Extension Relay API skill.mjs:writeNoteFile 写入 Markdown 文件 Suspicious artifacts and egress
http://127.0.0.1:3459 SKILL.md:17
https://clawhub.ai/user/fatmind skill-card.md:9
https://clawhub.ai/fatmind/skills/xhs-fav-export skill-card.md:39
https://clawhub.ai/fatmind/skills/webclaw3-browser-automation skill-card.md:40
https://www.xiaohongshu.com/ skill-card.md:41
https://www.xiaohongshu.com/user/profile/$ skill.mjs:388
Dependencies and supply chain
There are no structured dependency warnings.
File composition
skill.mjs wc3-code.mjs skill-card.md SKILL.md