illumio

Illumio MCP Server provides programmatic access to an Illumio Policy Compute Engine (PCE) via an interactive command interface. It exposes tools to manage workloads and labels, analyze traffic flows, and check PCE health. Users can create, update, and delete workloads and labels, retrieve traffic flow data with rich filtering options, and perform policy-related queries such as getting rulesets or IP lists. The server is designed to be driven by natural language prompts and structured tool commands, enabling automation and integration with other systems. To use it, configure the PCE connection details as environment variables, start the uv-based server, and then issue tool commands such as get-workloads, create-workload, get-traffic-flows, or check-pce-connection to validate connectivity.

Prerequisites:

• Python 3.8+
• Git
• Access to an Illumio PCE with valid API credentials

Installation steps:

1. Clone the repository: git clone [repository-url] cd illumio-mcp

2. Install dependencies: pip install -r requirements.txt

3. Set up environment variables (optional for local testing): export PCE_HOST=your-pce-host export PCE_PORT=your-pce-port export PCE_ORG_ID=1 export API_KEY=your-api-key export API_SECRET=your-api-secret

4. Run the MCP server using uv as described in the Configuration section: uv --directory /path/to/illumio-mcp run illumio-mcp

Tips:

• Ensure your Illumio PCE credentials have the necessary permissions for workloads, labels, and traffic data.
• If you modify configuration or environment, restart the uv process to apply changes.
• The server includes comprehensive error handling; issues will surface with detailed logs. Enable DEBUG logs for troubleshooting if needed.
• When using Claude Desktop or similar tooling, ensure the mcpServers.custom_settings block is correctly placed so the client can discover and start the server.
• You can extend or customize tool capabilities by editing the corresponding script/module that implements the Illumio APIs (workloads, labels, traffic, etc.).