veritasgraph-mcp

VeritasGraph MCP Server

mcp-name: io.github.bibinprathap/veritasgraph

The first zero-trust, air-gapped Enterprise GraphRAG server for the Model Context Protocol.

Connect your local IDE agent — Claude Desktop, Cursor, VS Code, Windsurf, Cline, Continue — directly to the VeritasGraph knowledge-graph engine over a standard JSON-RPC 2.0 stdio stream. Build graphs from your documents, run multi-hop graph-grounded queries with verifiable [doc#chunk] citations, and inspect entities — all 100% locally, with zero external data egress.

• Zero-trust / air-gapped — stdlib-only JSON-RPC server; no cloud calls, no telemetry. The only network hop is your local Ollama runtime.
• Verifiable attribution — every node and edge records the source chunk it came from, so answers cite their evidence.
• Drop-in — wraps the same studio_api.graphrag_engine used by Studio.

Tools

Tool	Purpose
veritasgraph_ingest_document	Chunk a document, extract entities/relationships, merge into the graph.
veritasgraph_query	Multi-hop, graph-grounded answer with citations and reasoning path.
veritasgraph_search_entities	Fast subgraph retrieval for a query (no LLM call).
veritasgraph_get_graph	Return the full graph: nodes, edges, stats.
veritasgraph_clear_graph	Clear the entire graph (destructive).

Resources

URI	Description
veritasgraph://graph	Live snapshot of the knowledge graph.
veritasgraph://stats	Entity / relationship / source counts.

Requirements

• Python 3.10+

• VeritasGraph repo dependencies: pip install -r requirements.txt

• A local Ollama runtime with a chat model pulled (ingest/query need a model; search_entities/get_graph do not):

  ollama serve & ollama pull qwen3:latest
  

Run it

# From the repository root
python -m veritasgraph_mcp            # stdio JSON-RPC server
python -m veritasgraph_mcp --debug    # verbose logging on stderr

Environment variables

Variable	Default	Meaning
VERITASGRAPH_MODEL	qwen3:latest	Default Ollama model for ingest/query.
OLLAMA_HOST	127.0.0.1:11434	Local Ollama endpoint.
STUDIO_DATA_DIR	studio_api/data	Where the graph snapshot is persisted.

Use it through VeritasGraph Studio

The same tools and resources are exposed over HTTP by the Studio backend, so you can drive the MCP server without an IDE — and the Studio agent pipeline can call its tools directly.

Start the Studio API and the bridge mounts under /mcp:

uvicorn studio_api.main:app --port 8200

Method & path	Purpose
GET /mcp/	Server identity, capabilities, tool/resource counts.
GET /mcp/tools	List tool definitions (same schema as tools/list).
POST /mcp/tools/{name}	Invoke one tool; body is the tool arguments.
GET /mcp/resources	List veritasgraph:// resources.
GET /mcp/resources/read?uri=...	Read a resource.
POST /mcp/rpc	Full JSON-RPC 2.0 bridge (any MCP method over HTTP).

Two ready-to-use tools are pre-registered in the Studio Tools section (VeritasGraph MCP · Query and VeritasGraph MCP · Search), pointing at loopback /mcp/tools/... endpoints. Enable Tools on an agent to have the orchestrator invoke them automatically during a run.

# Example: call a tool via the bridge
curl -s http://127.0.0.1:8200/mcp/tools/veritasgraph_search_entities \
  -H 'Content-Type: application/json' -d '{"query": "acme corp"}'

Register in your IDE

Replace /abs/path/to/VeritasGraph with the absolute path to your clone, and point command at the repo's Python (e.g. .venv/bin/python).

Claude Desktop — claude_desktop_config.json

{
  "mcpServers": {
    "veritasgraph": {
      "command": "/abs/path/to/VeritasGraph/.venv/bin/python",
      "args": ["-m", "veritasgraph_mcp"],
      "cwd": "/abs/path/to/VeritasGraph",
      "env": { "VERITASGRAPH_MODEL": "qwen3:latest" }
    }
  }
}

Cursor — .cursor/mcp.json

{
  "mcpServers": {
    "veritasgraph": {
      "command": "/abs/path/to/VeritasGraph/.venv/bin/python",
      "args": ["-m", "veritasgraph_mcp"],
      "cwd": "/abs/path/to/VeritasGraph"
    }
  }
}

VS Code — .vscode/mcp.json

{
  "servers": {
    "veritasgraph": {
      "type": "stdio",
      "command": "/abs/path/to/VeritasGraph/.venv/bin/python",
      "args": ["-m", "veritasgraph_mcp"],
      "cwd": "/abs/path/to/VeritasGraph"
    }
  }
}

Windsurf, Cline, and Continue use the same command / args / cwd shape.

Quick manual test (no IDE)

Pipe raw JSON-RPC frames over stdio:

printf '%s\n' \
  '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{}}' \
  '{"jsonrpc":"2.0","id":2,"method":"tools/list","params":{}}' \
  '{"jsonrpc":"2.0","id":3,"method":"tools/call","params":{"name":"veritasgraph_get_graph","arguments":{}}}' \
  | python -m veritasgraph_mcp

You should get three JSON-RPC responses: server info, the tool catalog, and the current graph snapshot.

How it fits VeritasGraph

The MCP server is a thin protocol adapter. All graph construction, multi-hop retrieval, and citation logic live in studio_api/graphrag_engine.py, the same engine that powers Studio. That means the MCP surface, the Studio UI, and the HTTP API all read and write the same local knowledge graph.