Frequently Asked Questions about Port MCP Server

Q: What is the Port MCP Server? A: The Port MCP Server is a Model Context Protocol (MCP) server that allows AI models and applications to interact with external data sources and tools using natural language. It enhances AI’s ability to understand and respond to context by providing real-time data and system insights.

Q: What can I do with the Port MCP Server? A: You can use it to quickly find information about your systems (e.g., entity details, on-call status), analyze scorecards to identify weak points, and create resources such as scorecards and rules to improve quality and compliance.

Q: What are the prerequisites for installing the Port MCP Server? A: You need a Port account with valid credentials (Client ID and Client Secret), and either Docker or uvx installed on your system.

Q: How can I install the Port MCP Server? A: You can install it using either Docker (via docker pull ghcr.io/port-labs/port-mcp-server:latest) or uvx (using the mcp-server-port package).

Q: How do I configure the Port MCP Server to work with my tools (Claude, Cursor, VS Code, Neovim)? A: You need to edit the configuration file for each tool and add the Port MCP Server configuration, including the command to run the server, the necessary arguments, and environment variables with your Port credentials. Detailed instructions are provided for each tool in the documentation.

Q: What is the purpose of the Client ID and Client Secret? A: The Client ID and Client Secret are your Port account credentials. They are required to authenticate your MCP server and allow it to securely access your Port data.

Q: What is the Port Region and why is it important? A: The Port Region specifies the geographical location of your Port account (either EU or US). Setting the correct region ensures that your MCP server connects to the correct Port API endpoint.

Q: What are the available tools in the Port MCP Server? A: The server provides a range of tools, including Blueprint Tools (e.g., get_blueprints, create_blueprint), Entity Tools (e.g., get_entities, create_entity), Scorecard Tools (e.g., get_scorecards, create_scorecard), and an AI Agents Tool (invoke_ai_agent).

Q: Can I use the Port MCP Server for local development? A: Yes, you can configure your MCP client to use a local cloned repository for testing and development purposes. You need to set up a virtual environment and configure the command and PYTHONPATH settings accordingly.

Q: Where can I provide feedback or request new features? A: You can share your feedback and feature requests on the Port roadmap page: https://roadmap.getport.io/ideas.

Q: What license is the Port MCP Server released under? A: The Port MCP Server is licensed under the MIT License, allowing you to use, modify, and distribute the software freely, subject to the terms and conditions of the license.