RAGScore

RAGScore is an MCP server designed to automate the generation of question-answer datasets and quantitative evaluation of Retrieval-Augmented Generation (RAG) systems using any LLM. It supports privacy-first workflows, runs locally or in the cloud, and provides instant evaluations, detailed multi-metric reports, and visualizations. It integrates seamlessly with tools like Ollama for local LLMs, Jupyter/Colab for notebooks, and production environments using a CLI or Python API, making it ideal for AI developers, ML engineers, and RAG system builders who need fast, reliable, and private evaluation of information retrieval pipelines.