root-signals-mcp

root-signals-mcp

5
ai_chatbotdeveloper_tools

Root Signals MCP Server is a Model Context Protocol server providing tools for AI assistants to evaluate responses based on quality. It supports various evaluations and has compatibility with multiple MCP clients like Cursor.

Root Signals MCP Server

This project serves as a bridge between Root Signals API and Model Context Protocol (MCP) client applications, enabling AI assistants and agents to evaluate responses against quality criteria.

Features

  • Exposes Root Signals evaluators as MCP tools.
  • Supports standard evaluation and RAG evaluation with contexts.
  • Implements SSE for network deployment.
  • Compatible with various MCP clients such as Cursor.

How to use this server

  1. Get your API key from Root Signals.
  2. Run the MCP Server with SSE transport on Docker.

Tools

The server exposes several tools like listing evaluators, running evaluations by ID or name, conducting RAG evaluations, and running coding policy adherence checks.

Usage Examples

  1. Evaluate and improve explanations with Cursor Agent.
  2. Use the MCP reference client directly from code.
  3. Measure prompt templates in Cursor.

How to Contribute

Contributions should be suitable for all users. Use Docker and pytest for testing before submission.

Limitations

  • Current implementation lacks backoff and retry mechanisms.
  • Bundled MCP client is for reference only.