omni-mcp
Pass
Audited by Gen Agent Trust Hub on Aug 13, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation references the
omniroutepackage to be run vianpxfrom the NPM registry. This package is maintained by the skill author and provides the necessary logic for the server connection. - [COMMAND_EXECUTION]: The provided configuration examples for Claude Desktop include shell commands that use
npxto execute the vendor's package. This is standard for setting up the described tool. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data through various tools like
memory_addandomniroute_route_request, which could potentially be manipulated by external inputs. - Ingestion points: Tools including
memory_addandomniroute_route_requestas defined in the skill markdown. - Boundary markers: None identified within the skill's instructions.
- Capability inventory: A set of 37 tools spanning health checks, routing, memory management, and administration.
- Sanitization: Not specified in the provided skill content.
Audit Metadata