fact-check
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external web sources, which is an inherent part of its fact-checking function.
- Ingestion points: External data is ingested through the
web_searchandfetch_urltools while verifying claims. - Boundary markers: The instructions do not define specific delimiters to isolate external content from the agent's core logic, though the focus is limited to verifying specific factual shortlists.
- Capability inventory: The skill can modify course content using the
pro-editingtool, but this action is gated behind a mandatory user approval step (ask_user). - Sanitization: There is no mention of filtering or sanitizing the content retrieved from external URLs.
Audit Metadata