mcp-server-port 0.1.4

A Model Context Protocol (MCP) server for Port.io

Port MCP Server

A Model Context Protocol (MCP) server for the Port.io API, enabling Claude to interact with Port.io's developer platform capabilities.

Installation

Obtain your Port credentials

1. Create a Port.io Account:

   • Visit Port.io
   • Sign up for an account if you don't have one

2. Create an API Key:

   • Navigate to your Port.io dashboard
   • Go to Settings > Credentials
   • Save both the Client ID and Client Secret

Claude Desktop

Add the following to your claude_desktop_config.json:

{
  "mcpServers": {
    "port": {
      "command": "uvx",
      "args": [
        "mcp-server-port@0.1.4",
        "--client-id", "YOUR_CLIENT_ID",
        "--client-secret", "YOUR_CLIENT_SECRET",
        "--region", "REGION" # US or EU
      ]
    }
  }
} 

Cursor

1. Make sure uvx is installed:

pip install uvx

2. Get its location:

which uvx
# Example output: /Users/janedoe/.local/bin/uvx

3. Create a script to run the server:

# run-port-mcp.sh

cd /Users/janedoe/.local/bin/uvx

# Run the server with the specified credentials
./.venv/bin/uvx mcp-server-port@0.1.4 --client-id YOUR_CLIENT_ID --client-secret YOUR_CLIENT_SECRET --region YOUR_REGION

4. Make it executable:

chmod +x /path/to/your/file/run-port-mcp.sh

5. Configure in Cursor settings:
   • Go to Cursor settings > MCP Servers
   • Configure with:
     • Name - Port
     • Type - Command
     • Command - /path/to/your/file/run-port-mcp.sh

Capabilities

Agent Tools

1. trigger_port_agent
   • Trigger the Port.io AI agent with a prompt and wait for completion
   • Required inputs:
     • prompt (string): The prompt to send to the Port.io AI agent
   • Returns: Agent response with status, output, and any required actions
   • Note: The agent may return action URLs for bug reports or other tasks that require user interaction

Blueprint Tools

1. get_blueprints

   • Retrieve a list of all blueprints from Port
   • Optional inputs:
     • detailed (boolean, default: false): Return complete schema details for each blueprint
   • Returns: Formatted text representation of all available blueprints

2. get_blueprint

   • Retrieve information about a specific blueprint by its identifier
   • Required inputs:
     • blueprint_identifier (string): The unique identifier of the blueprint to retrieve
   • Optional inputs:
     • detailed (boolean, default: true): Return complete schema details
   • Returns: Formatted text representation of the specified blueprint

Entity Tools

Will be added in the future.

Development

Local Setup

1. Create and activate a virtual environment:

# Create and activate virtual environment
python -m venv .venv
source .venv/bin/activate  # On Unix/macOS
# or
.venv\Scripts\activate  # On Windows

2. Install dependencies:

# Install in development mode
pip install -e .

# Install development dependencies
pip install -r requirements-dev.txt

3. Run the server locally:

python -m src.mcp_server_port --client-id "CLIENT_ID" --client-secret "CLIENT_SECRET" --region "REGION"

Publishing a New Version

To build and publish a new version to PyPI:

# Ensure you have the latest build tools
pip install --upgrade build twine

# Update version in pyproject.toml

# Build the package
python -m build

# Check the package
twine check dist/*

# Upload to PyPI
twine upload dist/mcp_server_port-0.1.4*

Troubleshooting

If you encounter authentication errors, verify that:

1. Your Port credentials are correctly set in the arguments
2. You have the necessary permissions
3. The credentials are properly copied to your configuration

License

This MCP server is licensed under the MIT License. This means you are free to use, modify, and distribute the software, subject to the terms and conditions of the MIT License. For more details, please see the LICENSE file in the project repository.