Evaluate RAG retrieval, agent tool calls and prompt regressions with deterministic metrics.
Free
2 servers across ai & agents and security & testing. The largest is Evalsmith.
AI & Agents · Security & Testing · github.com/diegoperezg7
Evaluate RAG retrieval, agent tool calls and prompt regressions with deterministic metrics.
Security scanner for MCP servers: tool poisoning, shadowing, permissions, secrets.
The rest either hold an API key issued by the publisher or ask for nothing.
Listed without a verification tick. Read the listing before installing.