Uma
AI & Memory
Local
A local RAG context optimizer MCP server that filters retrieved context to the minimum an LLM needs to answer correctly, using a cross-encoder and optional calibration.
How to connect
1
Glama registry
View https://glama.ai/mcp/servers/eypgr8kvbn for deploy options, or install from https://github.com/sainellutla/uma (see README for MCP config).
2
GitHub
Install from https://github.com/sainellutla/uma and add the server to your MCP client configuration (see repository README).
Tools
Tool names are not listed in our registry for this server. Connect it in your MCP client to see the live tool list.