CustomerGPT MCP
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@CustomerGPT MCPcreate a support bot and train it on my website"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
CustomerGPT MCP
Connect AI assistants to CustomerGPT over the Model Context Protocol. Let Claude, ChatGPT, Cursor or VS Code create support bots, train them on your website, read conversations and leads, and get the install snippet.
Option 1: hosted endpoint (recommended)
If your client supports remote MCP servers, add this URL and sign in with OAuth:
https://api.customergpt.ai/api/mcpNo install required. Anonymous demo tools are available at https://api.customergpt.ai/api/mcp/public.
Related MCP server: AI-Assisted CRM MCP Server
Option 2: local launcher
For clients that only start local (stdio) servers:
{
"mcpServers": {
"customergpt": {
"command": "npx",
"args": ["-y", "@customergpt/mcp"],
"env": { "CUSTOMERGPT_API_KEY": "cgpt_..." }
}
}
}The launcher forwards every request to the hosted endpoint, so new tools appear without updating the package. Requires Node.js 20+.
Variable | Description |
| Workspace API key. Without it only the anonymous demo tools are listed. |
| Backend origin. Default |
Prefer browser sign-in over an API key? Use @customergpt/cli: run customergpt login, then use customergpt mcp as the command.
Tools
Chatbots (list, get, create, update), knowledge sources (list, add, resync, delete), training jobs, test messages, conversations and leads, analytics, account usage, the install snippet and anonymous onboarding. Tools that change data ask the assistant to run a dry run and get your approval first.
Related
@customergpt/sdk— typed JavaScript/TypeScript client@customergpt/cli— command-line interface
License
MIT
Related MCP Connectors
Give AI agents secure access to RevDesk calling, SMS, phone numbers, caller IDs, and usage.
Manage AI assistants, history, calls, campaigns, contacts, knowledge, messaging, and automations.
Set up and run an in-product AI assistant: widgets, knowledge, MCP connections, usage.
Build and manage AI-native customer support agents from Claude or any MCP client.
Related MCP Servers
- AlicenseAqualityAmaintenanceEnables AI-powered ticket analysis and comprehensive Zendesk API integration via the Model Context Protocol, supporting both API token (CLI) and OAuth 2.1 (web) authentication modes.55522 npm5MIT
- FlicenseNot gradedqualityDmaintenanceEnables AI agents to perform CRM operations like creating contacts, managing deals, and updating leads through natural language using the Model Context Protocol.4-
- AlicenseNot gradedqualityDmaintenanceEnables AI clients to seamlessly take HubSpot actions and interact with your HubSpot data through the Model Context Protocol.29,792 npmMIT
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to manage CRM prospects, activities, reminders, and mailbox via natural language commands through the Model Context Protocol.3MIT