age-mcp

MCP server for Apache AGE graph databases.

Lets AI assistants (Claude, ChatGPT, Copilot) query and mutate graph data via the Model Context Protocol. Built with F# on .NET 10 – 1,300 lines, 21 tools, zero Python dependencies.

Get Started View on GitHub

What is age-mcp?

age-mcp is an MCP server that gives AI assistants direct access to Apache AGE graph databases running on PostgreSQL. Instead of writing Cypher queries manually, you describe what you want in natural language and the AI calls the right tools automatically.

Features

21 MCP tools – graph CRUD, Cypher queries, search, export/import, visualization, semantic search, Graph RAG
Data-compatible with agemcp (Python) – same tenant prefix, same property schema, no migration needed
Sub-millisecond metadata queries – cached list_graphs and get_schema respond in ~100 ns
Transactional batch operations – upsert_graph runs all mutations in a single transaction
Vector search – pgvector-backed semantic_search and graph_context (Graph RAG) with any OpenAI-compatible embedding API
Interactive visualization – generate_visualization produces standalone vis.js HTML
dotnet tool – install globally with dotnet tool install, no build required

Quick Start

# Start PostgreSQL + Apache AGE
docker compose up -d

# Install the MCP server
dotnet tool install --global AgeMcp

# Run
export AGE_CONNECTION_STRING="Host=localhost;Port=5435;Database=agemcp;Username=agemcp;Password=agemcp"
age-mcp

Then add to your Claude Desktop or Claude Code config:

{
  "mcpServers": {
    "age-mcp": {
      "type": "stdio",
      "command": "age-mcp",
      "env": {
        "AGE_CONNECTION_STRING": "Host=localhost;Port=5435;Database=agemcp;Username=agemcp;Password=agemcp",
        "TENANT_ID": "default"
      }
    }
  }
}

Use Cases

Knowledge graphs – build and query knowledge bases from AI conversations
People & org charts – model teams, roles, and relationships
Project tracking – link tasks, dependencies, and milestones in a graph
RAG with graph context – semantic search over vertices + N-hop neighbor expansion for richer LLM context
Data exploration – ask questions about existing graph data in natural language