claim-investigation
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external, untrusted content such as social media claims and viral narratives.
- Ingestion points: The skill accepts external claims from social media and other sources for systematic decomposition (Phase 1).
- Boundary markers: There are no explicit delimiters or specific 'ignore embedded instructions' warnings provided for the input claims.
- Capability inventory: The skill allows for file writing to the local filesystem (e.g., '{topic}-investigation-{date}.md') and performing external search queries.
- Sanitization: While the 'Claim Decomposition' methodology acts as a logical filter by breaking content into atomic facts, there is no technical sanitization or escaping of the input content before it is processed or written to files.
Audit Metadata