verify-factual-claim
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external factual claims and evidence, which creates a potential surface for indirect prompt injection attacks where malicious instructions could be hidden in the analyzed data.
- Ingestion points: The skill accepts user-provided factual claims and directs the agent to fetch evidence from authoritative specifications, documentation, and maintainer records (SKILL.md).
- Boundary markers: The instruction to "Rewrite the input as the narrowest falsifiable claim" may act as a structural delimiter, but there are no explicit security boundary markers or instructions to ignore embedded commands within the input.
- Capability inventory: The skill does not explicitly define or request high-privilege tools in its frontmatter, but the instructions imply the use of file reading and environmental observation tools to verify claims.
- Sanitization: There are no specific instructions for sanitizing, escaping, or filtering content retrieved from external evidence sources before it is processed by the agent.
Audit Metadata