Meetbot

Meetbot

Meet.bot is AI-native scheduling for people and agents. Check real calendar availability and book meetings on Google and Microsoft calendars through scheduling pages — and pay per meeting booked, not per user or seat. Actions: Book Meeting, Find Slots, Get Scheduling Page Info.

Category
访问服务器

README

Meet.bot MCP (Model Context Protocol)

A Model Context Protocol (MCP) server for the Meet.bot Booking Page API, enabling AI assistants to interact with scheduling and booking functionality.

What is this?

This MCP server connects AI assistants (like Claude, ChatGPT, and others that support MCP) to your Meet.bot account, allowing them to schedule meetings on your behalf. Instead of manually copying booking links or checking your calendar, you can simply ask your AI assistant to "schedule a 30-minute meeting with John next week" and it will handle the booking through your MeetBot scheduling pages.

How it works:

  1. Connect your MeetBot API token to the MCP server
  2. Configure your AI assistant to use this MCP server
  3. Ask your AI assistant to schedule meetings - it can check your availability, find time slots, and book meetings directly through your MeetBot account

This is particularly useful for busy professionals who want to automate meeting scheduling and let their AI assistant manage their calendar intelligently.

If you don't have an account, you can get one for free at https://meet.bot

Features

  • Complete API Coverage: Implements all endpoints from the Meet.bot Booking Page API v1
  • Type Safety: Full TypeScript support with comprehensive type definitions
  • Runtime Validation: Zod schemas for input validation and data integrity
  • Authentication Support: Bearer token authentication
  • Error Handling: Robust error handling with detailed error messages
  • Health Checks: Built-in health monitoring using the /v1/pages endpoint
  • MCP Protocol Compliance: Full Model Context Protocol server implementation
  • Dual Mode Support: Run locally (stdio) or remotely (HTTP/SSE)
  • Production Deployed: Live at https://mcp.meet.bot
  • Production Ready: Thoroughly tested and validated for production use

Installation

npm install @meetbot/mcp

Quick Start

1. Install and Configure

# Install the package
npm install @meetbot/mcp

# Or install globally for CLI usage
npm install -g @meetbot/mcp

2. Authentication

Authentication is not a tool—it is connection-based:

  • HTTP/SSE (remote, e.g. https://mcp.meet.bot): Send Authorization: Bearer <your_api_token> in the request headers when connecting. The server uses this token for all API calls in that session.
  • Stdio (local): Set the MEETBOT_AUTH_TOKEN environment variable when starting the server (e.g. MEETBOT_AUTH_TOKEN=your_token npx @meetbot/mcp).

There is no configure_meetbot or similar tool; the AI uses the tools below and auth is handled by the connection.

3. Use the Available Tools

The MCP server provides the following tools:

Get Scheduling Pages

await get_scheduling_pages();
// Returns all scheduling pages for the authenticated user

Get Page Information

await get_page_info({
  page: "https://meet.bot/user/30min"
});
// Returns detailed information about a specific scheduling page

Get Available Slots

await get_available_slots({
  page: "https://meet.bot/user/30min",
  count: 10,
  start: "2025-01-01",
  end: "2025-01-31",
  timezone: "America/New_York",
  booking_link: true
});
// Returns available booking slots with optional filters

Book a Meeting

await book_meeting({
  page: "https://meet.bot/user/30min",
  guest_email: "guest@example.com",
  guest_name: "Jane Doe",
  notes: "Meeting to discuss project requirements",
  start: "2025-01-15T14:00:00Z"
});
// Books a new meeting slot

Health Check

await health_check();
// Verifies API connectivity using the /v1/pages endpoint

Webhooks

Get notified the moment a meeting is booked, rescheduled, or cancelled. Meet.bot POSTs a JWT-signed (HS256) JSON payload — the same contract as the partner webhook — to your URL for each event (booking_received, booking_rescheduled, booking_cancelled).

// List your webhooks
await list_webhooks();

// Create (omit id) or update (pass id) a webhook
await set_webhook({
  webhook_url: "https://your-app.example.com/meetbot-hook",
  description: "CRM sync",
  coverage: "all",   // or "selected" with pages: [<page id>, ...]
  scope: "self"      // team admins can use "team" to also receive teammates' bookings
});
// Returns the webhook including the shared secret used to verify the signature.

// Delete a webhook
await delete_webhook({ id: 123 });

Deployment Options

The MCP server can be run in two modes:

1. Local Mode (Stdio Transport)

For local integration with AI assistants like Claude Desktop. Uses stdio transport for communication.

# Run locally
npx @meetbot/mcp

# Or with environment variable
MEETBOT_AUTH_TOKEN="your_token" npx @meetbot/mcp

2. HTTP Mode (SSE Transport)

For remote deployment with HTTP/SSE transport. This allows the MCP server to be accessed over the network.

Running the HTTP Server

# Build and start the HTTP server
npm run build
npm run start:http

# Or with custom port
PORT=8080 npm run start:http

# Or run directly with npx
npx meetbot-mcp-http

Server Endpoints

  • SSE Endpoint: GET /sse - Establishes an SSE connection for MCP communication
  • Messages Endpoint: POST /messages?sessionId=<id> - Receives client messages
  • Health Check: GET /health - Server health status

Authentication

All requests require a Bearer token in the Authorization header:

Authorization: Bearer <your-meetbot-api-token>

The same token is used to authenticate with the Meet.bot API.

Testing the HTTP Server

Local Testing:

# Health check (should fail without auth)
curl http://localhost:3000/health

# Health check with authentication
curl -H "Authorization: Bearer YOUR_TOKEN" http://localhost:3000/health

# Connect to SSE endpoint
curl -N -H "Authorization: Bearer YOUR_TOKEN" http://localhost:3000/sse

# Or use the test script
./test-http-server.sh YOUR_TOKEN

Production Testing:

# Test the live deployment
curl -H "Authorization: Bearer YOUR_TOKEN" \
     https://mcp.meet.bot/health

# Expected response:
# {"status":"ok","service":"meetbot-mcp"}

Deployment

The HTTP server can be deployed to various platforms:

  • Railway: railway upLive at 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 with Node.js directly

Example: Connecting to Production

# Your MCP client should connect to:
# https://mcp.meet.bot/sse

# With Authorization header:
# Authorization: Bearer <your-meetbot-api-token>

Example Dockerfile:

FROM node:18-alpine
WORKDIR /app
COPY package*.json ./
RUN npm ci --only=production
COPY dist ./dist
EXPOSE 3000
CMD ["node", "dist/cli-http.js"]

MCP Integration

Using with AI Assistants

This MCP server can be integrated with AI assistants like Claude, ChatGPT, and others that support the Model Context Protocol.

Configuration Example

{
  "mcpServers": {
    "meetbot": {
      "command": "npx",
      "args": ["@meetbot/mcp"],
      "env": {
        "MEETBOT_AUTH_TOKEN": "your_bearer_token_here"
      }
    }
  }
}

Available MCP Tools

The server exposes 8 tools for AI assistants:

  1. get_scheduling_pages - List all scheduling pages
  2. get_page_info - Get page details
  3. get_available_slots - Find available time slots
  4. book_meeting - Book a meeting
  5. health_check - Verify API connectivity using /v1/pages endpoint
  6. list_webhooks - List your outbound booking webhooks
  7. set_webhook - Create or update a webhook (booking_received / booking_rescheduled / booking_cancelled)
  8. delete_webhook - Delete a webhook

Authentication is provided by the connection (Bearer token in HTTP header for remote, or MEETBOT_AUTH_TOKEN for local stdio)—there is no separate configure tool.

Testing and Validation

The package has been thoroughly tested and validated:

MCP Protocol Compliance: Full JSON-RPC 2.0 support ✅ Tool Discovery: All 5 tools properly exposed ✅ Error Handling: Graceful error responses ✅ Type Safety: Complete TypeScript support ✅ Schema Validation: Input validation with Zod ✅ Production Ready: Tested with real MCP clients

Package Statistics

  • Package Size: 12.8 kB (58.0 kB unpacked)
  • Test Coverage: 21 passing tests
  • Dependencies: 3 production dependencies
  • TypeScript: 100% type coverage
  • Build Status: ✅ Passing
  • npm Registry: Published and available

Usage Examples

As a Library

import { MeetbotClient, MeetbotMCPServer } from '@meetbot/mcp';

// Use as a direct API client
const client = new MeetbotClient({
  authToken: 'your_token'
});

// Use as an MCP server
const server = new MeetbotMCPServer();
await server.run();

As an MCP Server

# Run the MCP server
npx @meetbot/mcp

# Or install globally
npm install -g @meetbot/mcp
meetbot-mcp

API Reference

MeetbotClient

The core API client for direct integration:

import { MeetbotClient } from '@meetbot/mcp';

const client = new MeetbotClient({
  authToken: 'your_token'
});

// Get all scheduling pages
const pages = await client.getPages();

// Get page information
const pageInfo = await client.getPageInfo({
  page: 'https://meet.bot/user/30min'
});

// Get available slots
const slots = await client.getSlots({
  page: 'https://meet.bot/user/30min',
  count: 20
});

// Book a meeting
const booking = await client.bookSlot({
  page: 'https://meet.bot/user/30min',
  guest_email: 'guest@example.com',
  guest_name: 'Jane Doe',
  start: '2025-01-15T14:00:00Z'
});

Data Types

All API responses are fully typed:

interface BookSlot {
  success: boolean;
  page: string;
  guest_email: string;
  guest_name: string;
  notes?: string;
  start: string;
  ical_uid: string;
}

interface PageInfo {
  title: string;
  duration: number;
  url: string;
  owner_name: string;
  max_days_into_the_future: number;
}

interface Slots {
  count: number;
  duration: number;
  slots: SlotDetails[];
}

Configuration

Environment Variables

You can configure the MCP server using environment variables:

export MEETBOT_AUTH_TOKEN="your_bearer_token"

# Then run the server
meetbot-mcp

Authentication

Auth is connection-based, not a tool:

  • HTTP/SSE: Send Authorization: Bearer <token> in request headers when connecting to the MCP server.
  • Stdio: Set MEETBOT_AUTH_TOKEN when running the server (e.g. MEETBOT_AUTH_TOKEN=your_token meetbot-mcp).

Development

Running Tests

npm test
npm run test:watch

Linting

npm run lint
npm run lint:fix

CLI Usage

The package includes a command-line interface for running the MCP server:

# Install globally
npm install -g @meetbot/mcp

# Run the MCP server
meetbot-mcp

# Or run directly with npx (no installation required)
npx @meetbot/mcp

Environment Variables

You can configure the server using environment variables:

export MEETBOT_AUTH_TOKEN="your_bearer_token"

# Then run the server
meetbot-mcp

Error Handling

The MCP server provides detailed error messages for common issues:

  • Configuration Errors: Missing or invalid configuration parameters
  • Authentication Errors: Invalid tokens
  • API Errors: Detailed error messages from the Meet.bot API
  • Validation Errors: Input validation failures with specific field errors

Contributing

  1. Fork the repository
  2. Create a feature branch
  3. Make your changes
  4. Add tests for new functionality
  5. Submit a pull request

License

MIT License - see LICENSE file for details.

Support

Changelog

1.2.10

  • Fix (stdio startup crash): the stdio server (npx @meetbot/mcp) declared a tools handler without advertising the tools capability, so the MCP SDK threw "Server does not support tools" and the process exited on launch. Now declares capabilities.tools; the server boots and answers initialize + tools/list (no token needed for introspection).

1.2.9

  • Server card: Added /.well-known/mcp/server-card.json for discovery and manual metadata (tools, prompts, authentication)
  • Smithery quality: Tool annotations (audience, priority), richer parameter descriptions, and prompts capability for quality scoring
  • Prompts (skills): Six MCP prompts for Smithery and clients: schedule_meeting, check_availability, book_for_guest, share_booking_link, list_my_pages, suggest_times with full prompts/list and prompts/get support

1.2.7

  • Example Config Update: Removed obsolete MEETBOT_BASE_URL from example configuration (base URL is hardcoded to https://meet.bot)

1.2.6

  • Documentation Updates: Updated production URL from Railway to https://mcp.meet.bot
  • Enhanced README: Added "What is this?" section explaining the MCP's purpose and how to use it for scheduling meetings through MeetBot accounts
  • User Onboarding: Added link to sign up for a free Meet.bot account

1.2.5

  • Jest Configuration: Updated Jest configuration for ES module support
  • Build Improvements: Excluded test files from TypeScript compilation
  • Publishing: Refined publishing checklist and documentation

1.1.1

  • ES Module Fix: Fixed ES module compatibility for Railway and other Node.js deployments
  • Import Extensions: Added proper .js extensions to all relative imports
  • Live Deployment: Successfully deployed to https://mcp.meet.bot
  • Production Validated: Confirmed working in production environment

1.1.0

  • HTTP/SSE Transport: Added HTTP server with SSE transport for remote deployment
  • Bearer Token Authentication: Implemented authentication for HTTP endpoints
  • Multi-mode Support: Can run as local stdio server or remote HTTP server
  • Deployment Ready: Added Dockerfile and deployment configurations for Railway, Fly.io, etc.
  • Health Check Endpoint: Added /health endpoint for monitoring
  • Session Management: Proper session handling with transport cleanup
  • Production Ready: Complete HTTP implementation with authentication and error handling

1.0.4

  • Fixed Health Check: Now uses the real /v1/pages endpoint instead of a fake health endpoint
  • Updated Documentation: Clarified health check implementation details
  • Improved Reliability: Health check now properly validates API connectivity

1.0.3

  • Simplified Configuration: Removed baseUrl requirement - now hardcoded to https://meet.bot
  • Updated Documentation: Simplified configuration examples
  • Improved UX: Users only need to provide authToken

1.0.2

  • Fixed Authentication: Removed unsupported sessionId parameter
  • Corrected API URL: Updated from api.meet.bot to meet.bot
  • Updated Documentation: Accurate authentication examples

1.0.1

  • Enhanced Documentation: Comprehensive README with usage examples
  • MCP Integration Guide: Added AI assistant configuration examples
  • Testing Results: Added validation and testing information

1.0.0

  • Initial release - Production-ready MCP server for Meet.bot API
  • Complete API coverage - All Meet.bot Booking Page API v1 endpoints
  • TypeScript support - Full type safety with comprehensive definitions
  • Runtime validation - Zod schemas for input validation and data integrity
  • MCP server implementation - Full Model Context Protocol compliance
  • CLI tool - Command-line interface for running the MCP server
  • Error handling - Robust error handling with detailed messages
  • Authentication - Support for bearer token authentication
  • Health checks - Built-in API connectivity monitoring using /v1/pages endpoint
  • Testing - Comprehensive test suite with 21 passing tests
  • Documentation - Complete API documentation and usage examples
  • Functional Testing - Verified with real MCP client requests
  • npm Publishing - Successfully published to npm registry

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