fact-check
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and analyze AI-generated content, which serves as an ingestion point for potentially untrusted data. This is the intended defensive purpose of the skill.
- Ingestion points:
SKILL.md(The 'Phase 1: Claim Extraction' section instructs the agent to extract and process statements from generated content). - Boundary markers: Absent (The instructions do not specify mandatory delimiters or 'ignore' headers for the input content being checked).
- Capability inventory:
SKILL.md(The 'Output Persistence' section indicates the agent has capabilities to write verification results and configuration to files such ascontext/output-config.mdand.fact-check-output.md). - Sanitization: Absent (The methodology focuses on verifying the truth of statements rather than sanitizing the input data for structural injection attacks).
- [NO_CODE]: The skill consists entirely of instructional Markdown and does not provide any scripts, executables, or automated tooling.
Audit Metadata