customercare-mcp-builder
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a development guide and does not execute any harmful code or commands. It provides standard reference implementations for web servers and MCP integration using well-known libraries.
- [SAFE]: The skill incorporates security-by-design principles, such as mandating the use of the 'audience' annotation to isolate sensitive data from the LLM and implementing a two-step 'prepare/confirm' pattern for irreversible transactions.
- [SAFE]: External dependencies referenced in the examples are standard industry libraries (e.g., Express, Starlette, Pydantic, Zod) used for their intended purposes.
- [SAFE]: Network operations described in the Knowledge/RAG integration examples are used to connect to user-configured OpenSearch instances and correctly utilize environment variables for credential management.
Audit Metadata