Pipedrive MCP Server

Pipedrive MCP Server

Provides comprehensive access to Pipedrive CRM with 100+ tools for managing deals, contacts, organizations, activities, and sales workflows through natural language conversations with Claude.

Category
访问服务器

README

Pipedrive MCP Server

npm version License: MIT Node Version

The most complete and robust Pipedrive MCP implementation for Claude

A production-ready Model Context Protocol server that provides Claude with comprehensive access to the Pipedrive CRM API. This server enables seamless automation of sales workflows, deal management, contact organization, and activity tracking through natural language conversations.

Features

  • 100+ Tools Across 10 Categories - Complete coverage of Pipedrive's core functionality
  • Advanced Rate Limiting - 10 requests/second with burst capacity up to 100 requests
  • Multi-Level Caching - 5-15 minute TTL for frequently accessed data
  • Retry Logic - Exponential backoff for failed requests (429, 500, 502, 503, 504)
  • Comprehensive Error Handling - Detailed error messages with actionable suggestions
  • Full TypeScript Support - Type-safe schemas and interfaces throughout
  • Zod Validation - Runtime validation for all inputs with helpful error messages
  • MCP Resources - Read-only access to pipelines, custom fields, and user info
  • MCP Prompts - 5 guided workflows for common operations
  • Performance Metrics - Built-in tracking for request duration and success rates
  • Read-Only Mode - Optional safety mode that blocks all write operations
  • Toolset Filtering - Enable/disable specific tool categories as needed

Tool Categories

Category Tools Description
Deals 23 Complete deal lifecycle management including creation, updates, stage movement, participants, products, and files
Persons 12 Contact management with custom fields, activities, deals, files, and follower management
Organizations 12 Company/organization management with relationships to persons, deals, and activities
Activities 8 Task, call, and meeting scheduling with due dates and completion tracking
Files 7 File upload, download, management, and remote file linking
Search 6 Universal search and entity-specific search across deals, persons, organizations, and products
Pipelines 8 Pipeline and stage management, including stage conversion statistics
Notes 5 Note creation and management for deals, persons, and organizations
Fields 8 Custom field discovery and metadata for all entity types
System 5 Health checks, metrics, user info, currencies, and cache management

Installation

Global Installation

npm install -g @iamsamuelfraga/mcp-pipedrive

Using npx (No Installation Required)

npx -y @iamsamuelfraga/mcp-pipedrive

Configuration

Prerequisites

  1. Get your Pipedrive API token from Settings > API
  2. Have Claude Desktop installed

Claude Desktop Setup

macOS

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

{
  "mcpServers": {
    "pipedrive": {
      "command": "npx",
      "args": ["-y", "@iamsamuelfraga/mcp-pipedrive"],
      "env": {
        "PIPEDRIVE_API_TOKEN": "your_api_token_here"
      }
    }
  }
}

Windows

Edit %APPDATA%\Claude\claude_desktop_config.json:

{
  "mcpServers": {
    "pipedrive": {
      "command": "npx",
      "args": ["-y", "@iamsamuelfraga/mcp-pipedrive"],
      "env": {
        "PIPEDRIVE_API_TOKEN": "your_api_token_here"
      }
    }
  }
}

Environment Variables

Variable Required Default Description
PIPEDRIVE_API_TOKEN Yes - Your Pipedrive API token
PIPEDRIVE_READ_ONLY No false Enable read-only mode (blocks all write operations)
PIPEDRIVE_TOOLSETS No deals,persons,organizations,activities Comma-separated list of enabled tool categories
LOG_LEVEL No info Logging level (debug, info, warn, error)

Advanced Configuration Examples

Read-Only Mode

Perfect for exploratory use or when you want to prevent accidental modifications:

{
  "mcpServers": {
    "pipedrive": {
      "command": "npx",
      "args": ["-y", "@iamsamuelfraga/mcp-pipedrive"],
      "env": {
        "PIPEDRIVE_API_TOKEN": "your_token",
        "PIPEDRIVE_READ_ONLY": "true"
      }
    }
  }
}

Filtered Toolsets

Only enable specific tool categories:

{
  "mcpServers": {
    "pipedrive": {
      "command": "npx",
      "args": ["-y", "@iamsamuelfraga/mcp-pipedrive"],
      "env": {
        "PIPEDRIVE_API_TOKEN": "your_token",
        "PIPEDRIVE_TOOLSETS": "deals,persons,search"
      }
    }
  }
}

Debug Logging

Enable verbose logging for troubleshooting:

{
  "mcpServers": {
    "pipedrive": {
      "command": "npx",
      "args": ["-y", "@iamsamuelfraga/mcp-pipedrive"],
      "env": {
        "PIPEDRIVE_API_TOKEN": "your_token",
        "LOG_LEVEL": "debug"
      }
    }
  }
}

Usage Examples

Example 1: Creating a Deal with Contact

Claude, create a new deal for "Enterprise Software License" worth $50,000.
The contact is John Smith (john@acme.com). Set the expected close date
to the end of next month and add a follow-up call for tomorrow.

Claude will:

  1. Search for or create the person "John Smith"
  2. Create the deal linked to this person
  3. Schedule a call activity for tomorrow
  4. Provide a summary with IDs and next steps

Example 2: Searching for Contacts

Find all contacts at Acme Corporation and show me their recent deals.

Claude will:

  1. Search organizations for "Acme Corporation"
  2. Get all persons associated with that organization
  3. Retrieve deals for each person
  4. Present organized results with totals

Example 3: Managing Activities

Show me all overdue activities for my open deals and reschedule them
to next week.

Claude will:

  1. List all activities with done=false and past due dates
  2. Filter for activities linked to open deals
  3. Update each activity with new dates next week
  4. Provide a summary of rescheduled items

Example 4: Using Custom Fields

Before creating this deal, show me what custom fields are available
for deals and explain what each one means.

Claude will:

  1. Access the pipedrive://custom-fields resource
  2. Extract deal-specific custom fields
  3. Display field names, types, and options
  4. Explain how to use them in deal creation

Example 5: Pipeline Management

Generate a pipeline report showing deal counts and total values for
each stage in my sales pipeline.

Claude will:

  1. Use the pipedrive://pipelines resource
  2. Get deal summaries grouped by stage
  3. Calculate totals and percentages
  4. Format as a readable report

Example 6: Weekly Review Workflow

Run the weekly pipeline review prompt.

Claude will:

  1. Execute the weekly-pipeline-review prompt
  2. Gather all open deals by stage
  3. Calculate metrics (won/lost, approaching close, stale deals)
  4. Generate actionable recommendations

Architecture

Core Components

  • PipedriveClient - HTTP client with rate limiting, caching, and retry logic
  • Rate Limiter - Bottleneck-based limiter (10 req/s, burst capacity)
  • Cache Layer - TTL-based cache with LRU eviction (500 item max)
  • Retry Handler - Exponential backoff for transient failures
  • Metrics Collector - Request tracking and performance monitoring
  • Error Handler - Standardized error formatting with context

Tool Structure

Each tool follows a consistent pattern:

  1. Zod Schema - Input validation with descriptive errors
  2. Description - Detailed usage instructions for the LLM
  3. Handler - Async function that calls PipedriveClient

Resources

Three MCP resources provide read-only reference data:

  • pipedrive://pipelines - All pipelines with stages and deal counts
  • pipedrive://custom-fields - Custom field definitions for all entities
  • pipedrive://current-user - Authenticated user info and permissions

Prompts

Five guided workflows for common operations:

  • create-deal-workflow - Complete deal creation with person and activity
  • sales-qualification - BANT qualification checklist
  • follow-up-sequence - Multi-day activity sequence
  • weekly-pipeline-review - Pipeline health report
  • lost-deal-analysis - Lost deal pattern analysis

Performance

Rate Limiting

  • Default: 10 requests/second (100ms between requests)
  • Burst: 100 token reservoir that refills every minute
  • Auto-retry: 429 errors automatically retry after 5 seconds

Caching Strategy

Data Type TTL Reason
Pipelines 10 min Pipeline structures change infrequently
Custom Fields 15 min Field definitions are relatively static
User Info 1 min User data may change during session
List Requests 5 min Default for paginated results

Metrics

The server tracks:

  • Total requests and success rate
  • Average response time
  • Error rate by type
  • Cache hit rate
  • Rate limit events

Access metrics with the system/metrics tool.

API Reference

This MCP server implements the Pipedrive REST API v1. For detailed API documentation, see:

Advanced Usage

Custom Field Discovery

Before creating or updating entities, check available custom fields:

// Access via MCP resource
pipedrive://custom-fields

// Or use field tools
fields/deal-fields
fields/person-fields
fields/org-fields
fields/activity-fields

Error Handling

All tools return structured errors with:

  • Error type (validation, authentication, rate limit, etc.)
  • Detailed message
  • Suggested actions
  • Original API error (if applicable)

Workflow Automation

Chain multiple tools together for complex workflows:

  1. Lead Qualification

    • Search for person
    • Get their deals and activities
    • Create qualification note
    • Update deal stage
  2. Deal Pipeline Movement

    • Get deal details
    • Check custom field requirements
    • Update custom fields
    • Move to next stage
    • Create next activity
  3. Reporting

    • List deals by stage
    • Get deal summaries
    • Calculate metrics
    • Format as markdown

Troubleshooting

See TROUBLESHOOTING.md for common issues and solutions.

Quick fixes:

  • Authentication errors: Verify your API token at https://app.pipedrive.com/settings/api
  • Rate limiting: Reduce request frequency or enable caching
  • Validation errors: Check tool input schema and required fields
  • Not seeing tools in Claude: Restart Claude Desktop after config changes

Contributing

We welcome contributions! Please see CONTRIBUTING.md for guidelines.

Development Setup

# Clone the repository
git clone https://github.com/iamsamuelfraga/mcp-pipedrive.git
cd mcp-pipedrive

# Install dependencies
npm install

# Build the project
npm run build

# Run tests
npm test

# Run with auto-reload during development
npm run dev

Running Tests

# Run all tests
npm test

# Run with coverage
npm run test:coverage

# Run with UI
npm run test:ui

Security

Please see SECURITY.md for our security policy and how to report vulnerabilities.

Important: Never commit your API token to version control. Always use environment variables.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Credits

Inspired by mcp-holded - an excellent MCP server implementation for Holded CRM.

Support

Changelog

See CHANGELOG.md for version history and release notes.

Roadmap

  • [ ] Webhook support for real-time updates
  • [ ] Bulk operations for mass updates
  • [ ] Advanced filtering with complex queries
  • [ ] Export/import functionality
  • [ ] Integration with other CRMs
  • [ ] GraphQL support

Made with dedication by Samuel Fraga

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选