compliance-officer

Pass

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION] (LOW): Identified an indirect prompt injection surface when processing external content. \n
  • Ingestion points: references/instructions.md directs the agent to fetch marketing copy and privacy policies via WebFetch. \n
  • Boundary markers: Absent. The instructions do not define clear delimiters or warning markers to help the agent distinguish between its instructions and the fetched untrusted content. \n
  • Capability inventory: The skill possesses network access (WebFetch) and file-read capabilities for local JSON rules. \n
  • Sanitization: No evidence of input validation or content filtering before processing.\n- [NO_CODE] (SAFE): This skill contains no executable scripts (e.g., Python, JavaScript, or Shell). It relies entirely on structured metadata, markdown instructions, and static JSON rule files.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 21, 2026, 06:14 AM