offer-honesty-check
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and audit external marketing copy, which could contain instructions meant to influence the agent's verdict or behavior.
- Ingestion points: The 'copy to check' provided by the user is the primary entry point for untrusted data into the agent's context.
- Boundary markers: The skill does not explicitly use delimiters or boundary markers to isolate the user-provided copy from the agent's instructions, though it does employ a structured auditing procedure.
- Capability inventory: The skill has the ability to execute a local Python script (scripts/flag_claims.py) to scan text and can rewrite the provided copy.
- Sanitization: No specific sanitization or filtering of the input text is described to prevent injection attacks.
Audit Metadata