llm-wiki-source-refresh
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of Markdown instructions and a reporting template; no executable scripts (Python, Node.js, or shell) are included in the package.
- [DATA_EXFILTRATION]: While the skill involves browsing external URLs and reading local files, these actions are limited to the stated purpose of source verification. The instructions explicitly include a safety gate that prohibits refreshing sensitive sources via external services without user approval.
- [PROMPT_INJECTION]: The skill ingests untrusted content from the web, which constitutes an indirect prompt injection surface. However, the risk is mitigated by explicit instructions to only propose patches for review, avoid silent rewrites, and maintain a human-in-the-loop workflow for contradictions.
Audit Metadata