gns3

This MCP server provides programmatic access to a complete GNS3 network simulation environment via the Model Context Protocol (MCP). It exposes a suite of tools categorized for server management, project handling, node control, topology operations, configuration, templates, snapshots, packet capture, drawing annotations, and advanced bulk operations. Use it to automate topology creation, manage projects and devices, push configurations, run bulk actions across multiple nodes, and gather status information from the GNS3 server. The tools are designed to work with the Gemini MCP client (or any MCP-compatible client) to send structured requests and receive consistent results.

With this server, you can list and manage GNS3 projects and devices, start/stop nodes, connect links, apply configuration templates, capture packets for troubleshooting, take and restore snapshots, and produce annotated topologies. The toolset is described in the repository docs, including references for each command, expected inputs, and outputs. For example, you can programmatically create a project, add devices from templates, start all devices, apply a network template, and then validate topology health using the provided validation tool. The 40+ tools enable AI-assisted network engineering workflows and bulk operations for enterprise-scale scenarios.

Prerequisites:

• A running GNS3 Server accessible at http://localhost:3080 (adjust GNS3_SERVER_URL as needed)
• Python 3.10+ installed
• A compatible MCP client environment (e.g., Gemini CLI) if you plan to integrate with Gemini

Step-by-step installation:

1. Clone the MCP server repository: git clone https://github.com/wael-rd/gns3-mcp-server.git cd gns3-mcp-server

2. Install the Python package in editable mode (from the repository root): python -m pip install -e .

3. Ensure the GNS3 server is reachable at the configured URL (http://localhost:3080 by default). If needed, set the environment variable GNS3_SERVER_URL to point to your GNS3 server.

4. Run the MCP server (as a local service or foreground process): python -m gns3_mcp_server

5. Optional: register with Gemini CLI or your MCP client of choice as described in the project docs: gemini mcp add gns3 "path/to/gns3-mcp-server/run.sh" # Linux/Mac gemini mcp add gns3 "path/to/gns3-mcp-server/run.bat" # Windows

6. Verify connectivity by issuing a simple MCP command through your client, for example listing projects or server info.

Prerequisites recap:

• Working GNS3 server and API access
• Python 3.10+ environment
• Network access between the MCP server and GNS3 server
• Client tooling (e.g., Gemini) if you want easy command registration and discovery

Notes and tips:

• Ensure GNS3 server URL is correctly configured via GNS3_SERVER_URL to avoid connection errors.
• The MCP server exposes many tools; refer to docs/TOOL_REFERENCE.md for full command syntax and examples.
• When running bulk operations, consider starting with a small scope (a single project) to validate behavior before applying changes across multiple devices.
• If you encounter authentication issues with GNS3, verify that the GNS3 server’s API is enabled and that any required API token or credentials are configured.
• For production deployments, consider running the MCP server as a background service and enabling proper logging and error monitoring.