Purpose
Meet.bot MCP is a Model Context Protocol server for the Meet.bot Booking Page API, used to connect AI assistants to a Meet.bot account so that the AI can schedule meetings on the user's behalf. The AI can check real calendar availability, find time slots, and complete bookings directly through the Meet.bot account, without manually copying booking links or checking the calendar. Its typical use case is described as busy professionals who want to automate meeting scheduling and have an AI assistant intelligently manage their calendar. If you do not yet have an account, you can register for free at https://meet.bot .
Main capabilities
- Full API coverage: implements all endpoints of the Meet.bot Booking Page API v1.
- Type safety: full TypeScript support with comprehensive type definitions.
- Runtime validation: uses Zod schema for input validation and data integrity checks.
- Authentication support: supports Bearer Token authentication.
- Error handling: robust error handling with detailed error information returned.
- Health check: built-in health monitoring via the /v1/pages endpoint.
- MCP protocol compliance: a complete Model Context Protocol server implementation.
- Dual-mode support: can run locally (stdio) or remotely (HTTP/SSE).
- Production deployment: already live at https://mcp.meet.bot , and is described as fully tested and validated for production use.
The server exposes 8 tools to AI assistants (the official text elsewhere in the testing and validation section also says "all 5 tools"; the two numbers are inconsistent, and the tools are listed here according to the tool list):
- get_scheduling_pages: list all scheduling pages
- get_page_info: get page details
- get_available_slots: find available time slots
- book_meeting: book a meeting
- health_check: verify API connectivity using the /v1/pages endpoint
- list_webhooks: list your outbound booking webhooks
- set_webhook: create or update a webhook (booking_received / booking_rescheduled / booking_cancelled)
- delete_webhook: delete a webhook
It also provides six MCP prompts (prompts/skills): schedule_meeting, check_availability, book_for_guest, share_booking_link, list_my_pages, suggest_times, and supports the full prompts/list and prompts/get.
Webhook notifications: instant notifications when a meeting is booked, rescheduled, or cancelled. Meet.bot pushes each event to your URL as a JWT-signed (HS256) JSON payload, using the same contract as partner webhooks; events include booking_received, booking_rescheduled, booking_cancelled.
How to use
Authentication is not a tool, but connection-based:
- HTTP/SSE (remote, e.g. https://mcp.meet.bot）：连接时在请求头中发送 Authorization: Bearer <your_api_token>. The server uses this token in all API calls for that session.
- Stdio (local): set the MEETBOT_AUTH_TOKEN environment variable when starting the server, for example MEETBOT_AUTH_TOKEN=your_token npx @meetbot/mcp. The official documentation explicitly states that there is no configuration tool such as configure_meetbot; the AI uses the above tools directly, and authentication is handled by the connection.
Basic flow:
- Connect the Meet.bot API token to the MCP server.
- Configure the AI assistant to use this MCP server.
- Have the AI assistant schedule meetings—it can check availability, find time slots, and book directly through the Meet.bot account.
Deployment options:
- Local mode (stdio transport): for local integration with AI assistants such as Claude Desktop, communicating via stdio.
- HTTP mode (SSE transport): for remote deployment.
- SSE endpoint: GET /sse, establishes the SSE connection for MCP communication
- Message endpoint: POST /messages?sessionId=<id>, receives client messages
- Health check: GET /health, returns server health status
- All requests require a Bearer token in the Authorization header, and this token is also used to authenticate the Meet.bot API.
The HTTP server can be deployed to a variety of platforms (the following are official examples): Railway (railway up, already online https://mcp.meet.bot）、Fly.io（fly launch && fly deploy), Google Cloud Run (container-based deployment), AWS App Runner (container-based deployment), any VPS (run directly with Node.js). The project includes an example Dockerfile.
Environment variables: the MCP server can be configured via environment variables; authentication is based on the connection rather than the tools: HTTP/SSE sends Authorization: Bearer <token> at connection time; Stdio sets MEETBOT_AUTH_TOKEN when running the server (for example MEETBOT_AUTH_TOKEN=your_token meetbot-mcp).
Package information (officially provided): package size 12.8 kB (58.0 kB unpacked), 21 passing tests, 3 production dependencies, 100% TypeScript type coverage, build status passing, published to the npm registry. A command-line interface for running the MCP server is also provided.
Error handling: detailed error messages are provided for common issues, including configuration errors (missing or invalid configuration parameters), authentication errors (invalid token), API errors (detailed error messages from the Meet.bot API), and validation errors (input validation failures that indicate the specific field error).
Use cases
Suitable for users who want an AI assistant to automatically complete meeting scheduling, check calendar availability time slots, and book meetings directly; supports sharing via remote HTTP/SSE deployment, or local stdio integration with desktop AI assistants.
Limitations and notes
- You must have a Meet.bot account and API token; authentication is completed through the connection (Bearer token or MEETBOT_AUTH_TOKEN environment variable), and there is no separate configuration tool.
- The official documentation states that billing is per booked meeting, not per seat; specific pricing, plans, and free quota are not yet provided.
- The base URL is hardcoded as https://meet.bot，不再需要 baseUrl configuration (the outdated MEETBOT_BASE_URL has been removed from the example configuration).
- The number of tools is stated inconsistently across different sections (8 and 5); this introduction lists them according to the tool list, and the actual project release shall prevail.
- Licensed under the MIT License. For more documentation, see https://docs.meet.bot。