rag-anythink-mcp

Overview

RAG-Anything MCP Server is a production-ready MCP implementation that combines knowledge graph queries with document processing and multimodal AI capabilities. It features multiple query modes (naive, local, global, hybrid, mix, bypass) for flexible information retrieval, automatic entity and relationship extraction from documents, and hybrid storage using Neo4j and PostgreSQL with pgvector. The server supports processing images, tables, and equations from PDFs, making it suitable for complex document analysis tasks.

Try asking AI

After installing, here are 3 things you can ask your AI assistant:

• ingest_document

  Ingest text or PDF documents

  Try:Add this research paper to the knowledge baseTry:Process this PDF document for information extraction
• query_knowledge_graph

  Query with multiple modes (naive, local, global, hybrid)

  Try:Find information about machine learning using hybrid searchTry:Search for topics related to renewable energy using global community search
• query_multimodal

  Query with images, tables, equations

  Try:Extract insights from the table in this PDFTry:Analyze the mathematical equation in this document
• process_document_file

  Process PDF files with multimodal extraction

  Try:Process this PDF file for text, images and tablesTry:Extract all content from this research document
• insert_content_list

  Insert pre-parsed content

  Try:Add this pre-parsed document to the knowledge graphTry:Insert this content list into the database
• delete_data

  Delete documents by ID

  Try:Remove document with ID 12345 from the knowledge baseTry:Delete this outdated research paper
• get_graph_statistics

  Get graph statistics

  Try:Show me statistics about the knowledge graphTry:How many nodes and relationships are in the graph?
• get_config_info

  Get configuration info

  Try:Show me the current configurationTry:What's the default query mode set to?

Comparable tools

Installation

# Clone the repository
git clone https://github.com/serkanyasr/rag-anythink-mcp.git
cd rag-anythink-mcp

# Copy environment template
cp .env.example .env

# Edit .env and set your OpenAI API key
# OPENAI_API_KEY=sk-...

# Start all services
docker-compose up -d

# View logs
docker-compose logs -f rag-mcp

{
  "mcpServers": {
    "rag-anything": {
      "command": "docker-compose",
      "args": ["up", "rag-mcp"],
      "env": {
        "OPENAI_API_KEY": "your-api-key"
      }
    }
  }
}

Compare rag-anythink-mcp with