wind-alice-meeting-expert
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads financial reports and meeting materials from Wind's official servers (
alice.wind.com.cn,mcp.wind.com.cn). These are vendor-owned resources used for the skill's primary function and do not pose a security risk. - [COMMAND_EXECUTION]: The skill uses
nodeandspawnto execute its internalrequest.jsscript and a daily update check script (update-check.mjs). These are standard operational patterns for a Node.js CLI tool. - [REMOTE_CODE_EXECUTION]: The
update-check.mjsscript facilitates skill updates by callingnpx skills updateornpx skills add. These operations target the official Wind repository on GitHub or Gitee, which are trusted sources for this specific vendor. - [DATA_EXPOSURE]: The skill reads the
WIND_API_KEYfrom local configuration files (~/.wind-aifinmarket/configorconfig.json). It contains explicit instructions and code logic to prevent these credentials from being printed in the output or leaked to the user. - [INDIRECT_PROMPT_INJECTION]: The skill ingests user input via the
--promptargument and sends it to an external AI service. While this is an ingestion point, the skill operates as a pass-through CLI, and the actual processing is handled by the vendor's backend, which typically has its own safety guardrails.
Audit Metadata