fact-check
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process arbitrary user-provided text to extract factual assertions, which creates a potential surface for indirect prompt injection.
- Ingestion points: Step 0 and Step 2 specify the processing of documents or blocks of text for claim extraction.
- Boundary markers: The instructions do not explicitly mandate delimiters for input data, though Step 1 (Falsifiability Test) serves as an inherent filter to separate factual claims from evaluative instructions.
- Capability inventory: The agent utilizes web search and browsing capabilities to verify information against third-party sources.
- Sanitization: No explicit sanitization or escaping of external content is defined, as the skill's primary function is to audit the content itself.
Audit Metadata