MCP Server

The Topolograph MCP Server exposes the Topolograph API through the Model Context Protocol, so LLM agents (Claude, and others) can query topologies, monitor events, and compute paths in real time — in natural language, without bespoke API glue.

vadims06/topolograph-mcp-server

Run it

The MCP server is bundled in topolograph-docker — the easiest way to run it as part of the full stack:

git clone https://github.com/Vadims06/topolograph-docker.git
cd topolograph-docker
docker-compose pull
docker-compose up -d

It comes up at http://localhost:8000/mcp and connects to the Topolograph API automatically.

Standalone

pip install -r requirements.txt
export TOPOLOGRAPH_API_BASE="https://your-topolograph-api-url"
export TOPOLOGRAPH_API_TOKEN="your-api-token"   # optional
python mcp-server.py

Default endpoint: http://0.0.0.0:8000/mcp.

Available tools

Tool	Purpose
get_all_graphs	List available graphs with filtering
get_graph_by_time	Fetch a specific graph by time
get_network_by_graph_time	Query network info (by IP, node ID, or mask)
get_graph_status	Check graph health and connectivity
get_network_events	Retrieve network up/down events
get_adjacency_events	Get node/host and link events
get_nodes	Query diagram nodes
get_edges	Query diagram edges
get_shortest_path	Compute shortest paths (with backup-path support)
upload_graph	Upload a new graph

What it unlocks

With these tools wired into an agent, you can ask questions like "which graphs are connected right now?", "what's the route between these two IPs?", or "what changed after the last topology event?" and get answers grounded in your real IGP — which is exactly what the AI Agent is built on.

---

Related: AI Agent · Python SDK · Quick Start with Docker