GoHighLevel MCP Server
A Model Context Protocol (MCP) server that provides tools for managing GoHighLevel (GHL) conversations, tasks, and calendar appointments through AI assistants like Claude.
README
GoHighLevel MCP Server
A Model Context Protocol (MCP) server that provides tools for managing GoHighLevel (GHL) conversations, tasks, and calendar appointments through AI assistants like Claude.
Features
Conversations
- search_conversations - Search and filter conversations with various criteria
- get_conversation - Get details of a specific conversation
- create_conversation - Create a new conversation with a contact
- update_conversation - Update conversation (star, assign, mark as read)
- delete_conversation - Delete a conversation
- get_messages - Get messages in a conversation
- send_message - Send SMS, Email, WhatsApp, or other message types
Tasks
- get_tasks - Get all tasks for a contact
- get_task - Get a specific task
- create_task - Create a new task
- update_task - Update an existing task
- delete_task - Delete a task
- complete_task - Mark a task as completed/incomplete
Calendar & Appointments
- get_calendars - Get all calendars in the location
- get_calendar - Get details of a specific calendar
- get_free_slots - Get available time slots
- get_calendar_events - Get events within a date range
- get_appointment - Get appointment details
- create_appointment - Create a new appointment
- update_appointment - Update an appointment
- delete_appointment - Delete an appointment
Prerequisites
- Node.js 18 or higher
- A GoHighLevel account with API access
- A Private Integration Token (PIT) from GoHighLevel
Getting Your GHL Credentials
- Log into your GoHighLevel sub-account
- Go to Settings > Integrations > Private Integrations
- Click Create New Integration
- Select the required scopes:
- Contacts: Read, Write
- Conversations: Read, Write
- Conversation Messages: Read, Write
- Calendars: Read, Write
- Calendar Events: Read, Write
- Copy the generated Private Integration Token
- Note your Location ID (found in Settings > Business Profile or in the URL)
Installation
# Clone or download this repository
cd ghl-mcp-server
# Install dependencies
npm install
# Build the project
npm run build
Configuration
Environment Variables
Set the following environment variables:
export GHL_API_KEY="pit-your-private-integration-token"
export GHL_LOCATION_ID="your-location-id"
Claude Desktop Configuration
Add the server to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"ghl": {
"command": "node",
"args": ["/absolute/path/to/ghl-mcp-server/dist/index.js"],
"env": {
"GHL_API_KEY": "pit-your-private-integration-token",
"GHL_LOCATION_ID": "your-location-id"
}
}
}
}
Cursor IDE Configuration
Add to your Cursor MCP settings:
{
"mcpServers": {
"ghl": {
"command": "node",
"args": ["/absolute/path/to/ghl-mcp-server/dist/index.js"],
"env": {
"GHL_API_KEY": "pit-your-private-integration-token",
"GHL_LOCATION_ID": "your-location-id"
}
}
}
}
Usage Examples
Once configured, you can use natural language to interact with your GHL account:
Conversations
- "Search for all unread conversations"
- "Get messages from conversation ID xyz123"
- "Send an SMS to contact abc456 saying 'Thank you for your inquiry!'"
- "Send an email to contact abc456 with subject 'Follow Up' and body 'Hi, just following up...'"
Tasks
- "Show me all tasks for contact xyz123"
- "Create a task for contact abc456: Call back tomorrow at 2pm"
- "Mark task xyz as completed"
- "Update task abc to change the due date to next Monday"
Calendar
- "List all calendars"
- "Show me appointments for the next 7 days"
- "Get free slots for calendar xyz between Jan 15 and Jan 20"
- "Create an appointment for contact abc456 on January 15th at 10am"
- "Cancel appointment xyz123"
API Reference
Conversation Tools
search_conversations
Search conversations with filters like contactId, assignedTo, query text, status, and message direction.
send_message
Send messages with support for:
- SMS: Simple text messages
- Email: With subject, HTML body, CC/BCC, attachments
- WhatsApp: WhatsApp messages
- IG/FB: Instagram and Facebook messages
- Live_Chat: Live chat messages
Task Tools
create_task
Create tasks with:
- Title (required)
- Description/body
- Due date (ISO 8601 format)
- Assignment to specific user
- Completion status
Calendar Tools
create_appointment
Create appointments with:
- Calendar ID (required)
- Contact ID (required)
- Start/end time (ISO 8601 format)
- Title and description
- Meeting location (Zoom, Google Meet, custom, etc.)
- Appointment status
- Notifications/automations toggle
Development
# Run in development mode
npm run dev
# Build only
npm run build
# Start production server
npm start
Troubleshooting
Common Issues
-
"GHL_API_KEY and GHL_LOCATION_ID environment variables are required"
- Ensure both environment variables are set correctly
-
"401 Unauthorized" errors
- Verify your Private Integration Token is valid
- Check that the token has the required scopes
-
"400 Bad Request" errors
- Verify the request parameters match the API requirements
- Check that IDs (contact, calendar, etc.) are valid
-
Server not appearing in Claude Desktop
- Verify the path to the server is correct
- Check the Claude Desktop logs for errors
- Restart Claude Desktop after configuration changes
Required Scopes
For full functionality, your Private Integration Token needs these scopes:
contacts.readonly- Read contact informationcontacts.write- Create and update tasksconversations.readonly- Read conversationsconversations.write- Create/update conversationsconversations/message.readonly- Read messagesconversations/message.write- Send messagescalendars.readonly- Read calendarscalendars.write- Manage calendarscalendars/events.readonly- Read eventscalendars/events.write- Create/update appointments
License
MIT
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。