Security audit
Security checks across malware telemetry and agentic risk
This skill is not clearly malicious, but it should be reviewed because it can steer an agent toward shell, file, and Feishu tools with loose scoping.
Install only if you intentionally want this skill to influence agent tool selection. Review or edit it so `exec` is only recommended for explicit weather or shell tasks, add stricter checks before `write` or Feishu management tools are suggested, and avoid the optional OpenClaw bundle patch unless you have audited the exact change and can roll it back.
- 文档 / create doc / 飞书文档 → use **feishu_doc** only when the user explicitly asks for a document
- 读文件 / read file → **read**; 写/编辑 → **write**
**Tool:** `get_recommended_tools({ user_message })` → returns `recommended_tools` (array of tool names) and `hint` (short instruction).60/60 vendors flagged this skill as clean.
No suspicious patterns detected.