topolograph

This MCP server provides programmatic access to the Topolograph API for analyzing OSPF and IS-IS network topologies. Through the MCP interface, you can query graphs, inspect nodes and edges, monitor events, and compute shortest paths with optional backup paths. Available tools include retrieval of all graphs, fetching graphs by time, querying network data within a graph, assessing graph health, and uploading new graphs. Use these capabilities to integrate network topology intelligence into AI agents, automate network analysis tasks, or augment decision-making with real-time topology data.

Prerequisites:

• Python 3.8+
• pip (comes with Python)
• Access to the Topolograph API and valid base URL (TOPOLOGRAPH_API_BASE)

Installation steps:

1. Clone or obtain the MCP server repository and navigate to its directory.
2. Install dependencies:

pip install -r requirements.txt

3. Configure environment variables (example):

export TOPOLOGRAPH_API_BASE="https://your-topolograph-api-url"
export TOPOLOGRAPH_API_TOKEN="your-api-token"  # optional

4. Run the MCP server:

python mcp-server.py

The server will start and listen by default at http://0.0.0.0:8000/mcp.

Docker/Docker-Compose users (optional): The Topolograph project mentions docker-compose integration. If you prefer using Docker, follow the topolograph-docker setup from the referenced repository to deploy the MCP server as part of the stack.

Notes and tips:

• The MCP server expects environment variables to configure the Topolograph API endpoint and optional authentication token. Adjust TOPOLOGRAPH_API_BASE and TOPOLOGRAPH_API_TOKEN as needed.
• By default, the server exposes the MCP endpoint at /mcp on port 8000. If you need a different port, run the server with a different configuration or proxy settings.
• If you encounter connectivity issues, verify that the Topolograph API is reachable from the host where the MCP server is running and that any required CORS or firewall rules permit access.
• The available tools include get_all_graphs, get_graph_by_time, get_network_by_graph_time, get_graph_status, get_network_events, get_adjacency_events, get_nodes, get_edges, get_shortest_path, and upload_graph. Use these to build rich AI-assisted network analytics workflows.