cvs-product-data
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill fetches product names, descriptions, and ingredients from CVS through a remote API. This untrusted external content is processed by the agent, creating a potential attack surface for indirect instructions.
- Ingestion points: Data retrieved from the search and product detail endpoints (
/api/v1/cvs/search,/api/v1/cvs/product) inSKILL.md. - Boundary markers: No specific boundary markers or instructions to treat the API output as untrusted are provided for the agent.
- Capability inventory: The skill uses
requestsand provides examples for network communication and script execution. - Sanitization: No sanitization or filtering logic is applied to the retrieved product data before it is presented to the agent.
- [EXTERNAL_DOWNLOADS]: The skill includes instructions to download and install vendor-related software packages and tools.
- Evidence: References to
pip install scavio==0.32.0,npm install scavio@0.32.0, andnpx -y @scavio/mcp-server@0.30.0in the SDK and setup sections ofSKILL.md.
Audit Metadata