b2c-custom-api-development
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for creating REST endpoints that ingest and process external data from HTTP request parameters and bodies, creating a potential surface for indirect prompt injection where malicious instructions in the input could influence agent behavior. \n
- Ingestion points: 'references/IMPLEMENTATION.md' demonstrates reading query parameters, path parameters, and request bodies using 'request.getHttpParameterMap()', 'request.getSCAPIPathParameters()', and 'request.httpParameterMap.requestBodyAsString'. \n
- Boundary markers: The provided implementation templates do not include boundary markers or safety instructions for the agent to ignore or delimit potentially malicious instructions within the ingested data. \n
- Capability inventory: The implementation code has access to the B2C Commerce Script API ('dw.system.*'), which provides capabilities for data processing, response rendering via 'RESTResponseMgr', and external service calls via 'LocalServiceRegistry'. \n
- Sanitization: The illustrative code examples do not demonstrate input validation or sanitization, representing a standard vulnerability surface that requires additional security implementation in production environments.
Audit Metadata