managing-controlled-substances
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external, untrusted data including PDMP reports, clinical assessments, and electronic prescriptions.
- Ingestion points: The 'Checkpoint A: Pre-Draft Intake' and 'Step 2: PDMP Review' sections require the agent to process external medical documents and reports.
- Boundary markers: The instructions do not define specific delimiters or instructions for the agent to ignore potentially malicious prompts embedded within these external documents.
- Capability inventory: The skill body is purely instructional markdown and does not utilize file-write, network communication, or subprocess execution tools, which significantly mitigates the potential impact of an injection.
- Sanitization: There are no explicit instructions for the agent to sanitize or validate the content of the ingested clinical data before processing it.
Audit Metadata