add-mcp-from-catalog
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a structured workflow for interacting with the Speakeasy AI Control Plane (AICP) platform. It follows security best practices by explicitly forbidding the agent from requesting sensitive information like API keys, passwords, or OAuth tokens from the user.
- [SAFE]: All critical actions, including project selection, candidate registration, and identity provider attachment, require explicit user confirmation and human-in-the-loop interactions for secure handoffs (OAuth/Secrets), which occur outside the chat environment.
- [SAFE]: The workflow relies on authenticated platform tools (
list_projects,search_mcp_catalog,register_catalog_mcp) rather than processing untrusted user-supplied URLs or data, minimizing the risk of prompt injection or unauthorized access.
Audit Metadata