fact-check
Pass
Audited by Gen Agent Trust Hub on Aug 23, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The instructions contain strict directives requiring the AI to adhere to the provided prompt verbatim and prohibit deviations from the specified workflow. These patterns are used here to maintain the structural integrity of the fact-checking process rather than to bypass safety controls.
- [INDIRECT_PROMPT_INJECTION]: The skill is intended to process untrusted data provided by the user for verification. * Ingestion points: User input placeholder for claims and instructions to process external documents, links, and chat logs in SKILL.md. * Boundary markers: No specific delimiters are used to isolate user-provided content from the skill's instructions. * Capability inventory: The skill instructions specify the use of network access for external verification. * Sanitization: The skill does not implement sanitization or validation for the input claims or external materials.
Audit Metadata