Skills
.Work. You
Rest

modular-rag-mcp-server

Official

Deploy a modular RAG MCP knowledge hub

Software Engineering#embeddings#rag#mcp#dashboard#hybrid search#vector store#rag evaluation
AuthorAradotso
Version1.0.0
Installs0

System Documentation

What problem does it solve?

It eliminates the friction of building and operating a Retrieval-Augmented Generation (RAG) system you can plug into AI assistants via Model Context Protocol (MCP), so questions can reliably pull from your documents instead of guessing.

Core Features & Use Cases

  • Ingestion Pipeline: Converts PDFs to Markdown, chunks content, generates embeddings, and builds a vector index with multimodal image captioning.
  • Hybrid Retrieval: Combines dense (semantic) + sparse (BM25) search using RRF fusion, with optional reranking for better relevance.
  • MCP Tools for Assistants: Exposes knowledge-hub operations like querying and collection/document introspection for Claude Desktop and other MCP clients.
  • Dashboard + Observability: Provides a Streamlit dashboard for ingestion and query tracking plus white-box tracing to debug performance regressions.
  • Evaluation with Ragas: Runs regression-style evaluation using Ragas metrics to validate improvements over time.

Quick Start

Run the MCP server and point your assistant’s MCP configuration to the project’s server entry so you can query your RAG collections immediately.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: modular-rag-mcp-server
Download link: https://github.com/Aradotso/mcp-skills/archive/main.zip#modular-rag-mcp-server

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.