rafaeljegundo/pipedrive-mcp-server

Pipedrive MCP Server

A Model Context Protocol (MCP) server that provides comprehensive access to Pipedrive CRM data. Fetch deals, activities, pipeline statistics, and get AI-powered insights about your sales pipeline.

Features

• 🔍 Deal Management - Fetch and filter deals by pipeline, stage, owner, and status
• 📊 Pipeline Analytics - Get comprehensive statistics and insights about your pipelines
• 📋 Activity Tracking - Access activities, tasks, and next steps for deals
• 🤖 AI Analysis (Optional) - Get actionable insights about deal health and priorities
• 🔐 Secure - API credentials stored safely in environment variables

Installation

1. Clone or download this repository:

git clone https://github.com/yourusername/pipedrive-mcp.git
cd pipedrive-mcp

2. Install dependencies:

npm install

3. Set up your environment variables:

cp .env.example .env

4. Edit .env with your Pipedrive API token:

PIPEDRIVE_API_TOKEN=your_pipedrive_api_token_here

# Optional: For AI analysis features
OPENAI_API_KEY=your_openai_api_key_here

Getting Your Pipedrive API Token

1. Log into your Pipedrive account
2. Go to Settings & Privacy → Personal Preferences → API
3. Copy your personal API token

Configuration for Claude Desktop

Add this server to your Claude Desktop configuration:

macOS

Edit ~/Library/Application Support/Claude/claude_desktop_config.json:

Windows

Edit %APPDATA%\Claude\claude_desktop_config.json:

Add the following to the mcpServers section:

{
  "mcpServers": {
    "pipedrive": {
      "command": "node",
      "args": ["/absolute/path/to/pipedrive-mcp/src/index.js"],
      "env": {
        "PIPEDRIVE_API_TOKEN": "your_token_here"
      }
    }
  }
}

Available Tools

pipedrive_get_deals

Fetch deals with various filters.

Parameters:

• status - Deal status: open, won, lost, deleted, all_not_deleted (default: open)
• pipeline_id - Pipeline ID to filter by
• stage_id - Stage ID to filter by
• user_id - Owner ID to filter by
• filter_id - Use a predefined filter
• limit - Number of deals to return (max 500, default: 100)
• sort - Sort field and direction (e.g., "value DESC")

pipedrive_get_deal

Get detailed information about a specific deal.

Parameters:

• deal_id - The ID of the deal (required)

pipedrive_get_activities

Fetch activities (tasks, calls, meetings, etc.).

Parameters:

• deal_id - Filter by deal ID
• user_id - Filter by user ID
• type - Activity status: all, done, undone (default: all)
• limit - Number of activities to return (default: 100)

pipedrive_get_pipelines

Get list of all pipelines in your Pipedrive account.

pipedrive_get_stages

Get stages for a specific pipeline.

Parameters:

• pipeline_id - Pipeline ID (defaults to pipeline 1)

pipedrive_get_pipeline_stats

Get comprehensive statistics for a pipeline including:

• Total deals and value
• Breakdown by stage and owner
• Activity metrics
• Deal probability analysis

Parameters:

• pipeline_id - Pipeline ID to analyze

pipedrive_analyze_deals

AI-powered analysis of deals for actionable insights (requires OpenAI API key).

Parameters:

• pipeline_id - Pipeline to analyze
• stage_id - Specific stage to analyze
• user_id - Analyze deals for specific owner
• limit - Maximum deals to analyze (default: 50)
• model - AI model to use (default: gpt-4o-mini)

Usage Examples

Once configured in Claude Desktop, you can ask Claude:

• "Show me all open deals in pipeline 1"
• "What deals need immediate attention?"
• "Get activities for deal #123"
• "Show pipeline statistics and identify bottlenecks"
• "Which deals are at risk of stalling?"
• "What are the next steps for high-value deals?"

Environment Variables

Development

Run the server locally for testing:

npm start

Run with file watching for development:

npm run dev

Troubleshooting

"Pipedrive API error"

• Check that your API token is valid
• Ensure you have access to the requested resources
• Verify pipeline and stage IDs exist

AI analysis not working

• Ensure OPENAI_API_KEY is set in your environment
• Check that you have API credits available
• Try a smaller limit parameter to reduce token usage

No deals returned

• Verify the pipeline_id exists (use pipedrive_get_pipelines first)
• Check that you have permission to view deals in that pipeline
• Try removing filters to see all available deals

Security Notes

• Never commit your .env file or expose API tokens
• Use environment variables or secure credential storage
• The server only has read access to Pipedrive data
• AI analysis is optional and processed locally

License

MIT

Contributing

Contributions are welcome! Please feel free to submit issues or pull requests.