8 min

MCP Configuration

Connect AI Product Manager to Claude Code, Cursor, and other AI-powered development tools using Model Context Protocol.

What is MCP?

Model Context Protocol (MCP) is an open standard that enables AI-powered development tools to access specialized services and data sources. By connecting AI Product Manager via MCP, you can seamlessly integrate product planning, architecture design, and task generation directly into your development workflow.

This integration allows Claude Code, Cursor, and other AI coding assistants to access your AI Product Manager capabilities without leaving your development environment.

Generate Your API Key

To connect AI Product Manager to any MCP-compatible tool, you'll need to generate an API key first.

Step 1: Access API Settings

  1. 1.Log in to your AI Product Manager account
  2. 2.Navigate to Settings → API Keys
  3. 3.Click "Generate New API Key"
  4. 4.Give your key a descriptive name (e.g., "Claude Code Integration")
  5. 5.Select the scopes you need (typically all scopes for full integration)
  6. 6.Click "Generate" and copy the key immediately

⚠️ Important: Your API key is sensitive. Keep it secure and never commit it to version control. If you suspect your key has been compromised, regenerate it immediately from the API Keys settings.

Connect with Claude Code

Claude Code is Anthropic's official CLI for Claude that enables AI-assisted development from your terminal.

Quick Setup

Open a terminal in your project directory and run the following command:

$ claude mcp add --transport http aipm-server "<AIPM_MCP_URL>" --header "Authorization:Bearer <YOUR_API_KEY>"

Replace the following:

  • →<AIPM_MCP_URL> with the MCP server endpoint provided by your AI Product Manager instance
  • →<YOUR_API_KEY> with the API key you generated in the previous step

✓ Done! Claude Code now has access to AI Product Manager capabilities. You can immediately start using it in your projects.

Using AI Product Manager in Claude Code

Once configured, you can use AI Product Manager capabilities by referencing them in Claude Code:

Generate a PRD:

Use AI Product Manager to generate a PRD for a social media platform

Create Architecture:

Generate system architecture for an e-commerce platform using AI Product Manager

Plan Sprints:

Create a 3-month sprint plan for building a mobile app with AI Product Manager

Connect with Cursor

Cursor is an AI-first code editor that integrates Claude and supports MCP for extended capabilities.

Step 1: Open Cursor Settings

  1. 1.Open Cursor and go to Settings (⌘, on Mac or Ctrl+, on Windows/Linux)
  2. 2.Search for "MCP" in the settings search
  3. 3.Click on "MCP Servers"

Step 2: Add AI Product Manager Server

  1. 1.Click "Add Server"
  2. 2.Enter the server name: aiProductManager
  3. 3.Enter the command: npx @artificial-outreach/ai-pm-mcp
  4. 4.Add environment variables:
API_KEY: your-api-key-here
API_URL: https://api.artificialoutreach.com
  1. 5.Click "Save" to confirm the configuration

Step 3: Use in Cursor

After adding the server, you can use AI Product Manager in Cursor's chat:

Type @ai-product-manager in the Cursor chat to access commands:

  • →Generate PRD - Create product requirement documents
  • →Design Architecture - Plan system architecture
  • →Create Roadmap - Plan sprints and timelines
  • →Generate Tasks - Create development tasks with descriptions

Environment Variables

The following environment variables can be configured when setting up the MCP server:

API_KEY

Your AI Product Manager API key (required). Generate this from your account settings.

API_URL

The API endpoint URL. Will be provided by your AI Product Manager instance.

WORKSPACE_ID

Optional. Specify a workspace ID to limit operations to a specific workspace. If not set, operations use your default workspace.

LOG_LEVEL

Optional. Set to debug, info, or error for debugging.

Troubleshooting

Connection Issues

If the MCP server won't connect:

  • →Verify your API key is correct and not expired
  • →Check that your network can reach the API URL
  • →Enable debug logging with LOG_LEVEL=debug
  • →Restart your editor after updating configuration

API Key Not Working

If you get authentication errors:

  • →Generate a new API key from your account settings
  • →Update the configuration with the new key
  • →Restart your editor or development tool

Next Steps

Ask me to learn more...