AI applicationsTools and integrations

Tools and integrations

Connect your AI application to Memgraph with the Memgraph MCP server, or use Memgraph from the AI framework you already work with.

  • Memgraph MCP server: Connect agents and LLM clients such as Claude to your Memgraph database over the Model Context Protocol (MCP).
  • LangChain: Build agentic applications with the Memgraph toolkit, and create and query knowledge graphs.
  • LlamaIndex: Create a knowledge graph from unstructured data and query it in natural language.
  • LightRAG: Use Memgraph as the graph storage for LightRAG’s retrieval-augmented generation.
  • Cognee: Extract concepts and relationships from text and search them as a knowledge graph.
  • Mem0: Store and query AI agent memory as a graph, with Mem0 1.x.

The diagram groups frameworks that work with Memgraph by what they do: knowledge graph creation, agent memory and agent frameworks.

Integrations grouped into knowledge graph creation (Graphiti, LightRAG), agentic memory (Mem0, Cognee) and agentic frameworks (MCP, LlamaIndex, LangChain, LangGraph, A2A), all connected to Memgraph

Memgraph Toolbox

The Memgraph Toolbox is a Python library of tools for querying and inspecting a Memgraph database, such as running Cypher and searching the graph schema. The Memgraph MCP server and the LangChain integration are built on it, and you can call the same tools from your own code or framework.

pip install memgraph-toolbox

The toolbox is part of the Memgraph AI Toolkit, which also includes the MCP server, the LangChain and LightRAG integrations, Unstructured2Graph, SQL2Graph and Context Graph.

Graph machine learningMemgraph MCP server