flexible-graphrag

Overview

Flexible GraphRAG is a comprehensive AI context platform that integrates with MCP to provide tools for ingesting documents from 13 data sources (9 with auto-sync), performing hybrid searches across vector, graph, and full-text databases, and conducting AI queries. The MCP server exposes the core functionality of the system to Claude Desktop and other MCP clients, allowing seamless integration into AI workflows. It supports multiple graph databases, vector stores, and LLM providers through a dual-framework architecture using LlamaIndex and LangChain.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• ingest_documents

  Ingest documents from various data sources into the knowledge graph

  Try:Ingest documents from the Alfresco repositoryTry:Upload documents from Amazon S3 bucketTry:Process documents from local filesystem directory
• search_documents

  Perform hybrid search across vector, full-text, and graph databases

  Try:Search for documents related to machine learning researchTry:Find documents containing information about customer contractsTry:Search using semantic similarity and keyword matching
• query_documents

  Query documents using natural language or SPARQL for RDF stores

  Try:What are the latest financial reports?Try:Find information about product development timelineTry:Query the knowledge graph for relationships between entities
• ingest_text

  Directly ingest text content into the knowledge graph

  Try:Ingest this text: 'Company policy requires all employees to complete annual training'Try:Add this document content to the knowledge baseTry:Process and index the following text snippet
• system_diagnostics

  Run system diagnostics and health checks

  Try:Check system health and performanceTry:Run diagnostic tests on the knowledge graphTry:Verify all database connections are working
• health_checks

  Perform health checks on system components

  Try:Check if the vector database is responsiveTry:Verify the graph database connection statusTry:Confirm all services are running normally
• get_processing_status

  Get the status of ongoing document processing tasks

  Try:Check status of the document ingestion jobTry:Show progress of knowledge graph extractionTry:List all active processing tasks
• cancel_processing_task

  Cancel an ongoing document processing task

  Try:Cancel the document ingestion jobTry:Stop the knowledge graph extraction processTry:Abort the text processing task
• list_data_sources

  List available data sources for document ingestion

  Try:Show all available data sourcesTry:List configured repositories and storage systemsTry:What data sources can I use for ingestion?

Comparable tools

Installation

{
  "mcpServers": {
    "flexible-graphrag": {
      "command": "python",
      "args": ["-m", "flexible_graphrag_mcp"]
    }
  }
}

FAQ

What data sources are supported by the MCP server?

The MCP server supports 13 data sources including filesystem, cloud storage (S3, Azure), enterprise repositories (Alfresco, SharePoint, Box, CMIS), and web sources. 9 of these support automatic incremental updates.

Can I use both LlamaIndex and LangChain frameworks with the MCP server?

Yes, the MCP server supports both LlamaIndex (default) and LangChain frameworks. Each pipeline stage can be configured independently to use either framework through environment variables.

Compare flexible-graphrag with