medicare-drug-stats
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves data from external federal healthcare datasets via the Socrata Open Data API. This creates a surface for indirect prompt injection if malicious instructions were present in the aggregate data or API metadata.
- Ingestion points: Centers for Medicare & Medicaid Services (CMS) data retrieved from data.cms.gov and Socrata API endpoints.
- Boundary markers: None explicitly defined in the prompt instructions to delimit the retrieved data from agent instructions.
- Capability inventory: The skill utilizes network access (curl/fetch) to query APIs and parses the resulting JSON data.
- Sanitization: No specific sanitization or validation logic for the API responses is detailed in the markdown instructions.
Audit Metadata