NetworkX

How to Set Up and Use NetworkX

The NetworkX MCP Server brings the full power of the Python NetworkX graph library into AI conversations, enabling Claude and other MCP clients to create, manipulate, and analyze graphs and networks without any additional cloud services. It supports core graph operations like node and edge management, path finding, and graph composition, as well as advanced analysis including centrality metrics, community detection, PageRank, maximum flow calculations, and PNG visualization generation. Researchers, data scientists, and developers working on network analysis, citation graphs, social networks, or infrastructure topology problems can use this server to run complex graph algorithms directly from natural language prompts.

pip install networkx-mcp-server

{
  "mcpServers": {
    "networkx": {
      "command": "python",
      "args": ["-m", "networkx_mcp"]
    }
  }
}

{
  "mcpServers": {
    "networkx": {
      "command": "python",
      "args": ["-m", "networkx_mcp"]
    }
  }
}

- "Create a graph called 'web', add nodes 1, 2, 3 with edges between them, then find the shortest path from 1 to 3"
- "Build a citation network with 5 papers and compute the PageRank of each node"
- "Detect communities in my social network graph and visualize the result"
- "Calculate betweenness centrality for all nodes in the 'infrastructure' graph"