Whatsapp MCP
A WhatsApp MCP (Model Context Protocol) server that connects your WhatsApp account with AI agents, enabling automated messaging, contact search, and chat interaction through LLMs.
WhatsApp MCP + CLI
A WhatsApp integration that works in two ways:
- MCP Server — lets AI assistants like Claude control WhatsApp directly
- CLI Tool — send/read WhatsApp messages from your terminal
Built on whatsapp-web.js.
Features
- Send WhatsApp messages by contact name or phone number
- Read recent messages from any chat
- List all recent chats with previews
- Search contacts by name
- Works as an MCP server (Claude, etc.) or standalone CLI
Setup
1. Install dependencies
npm install
2. Authenticate WhatsApp
node index.js auth
Scan the QR code with your WhatsApp mobile app. Your session is saved locally — you only need to do this once.
MCP Server (for Claude / AI assistants)
The MCP server exposes 4 tools over stdio:
| Tool | Description |
|---|---|
send_message | Send a message to a contact |
read_messages | Read recent messages from a chat |
list_chats | List recent chats with previews |
search_contacts | Search contacts by name |
Configure in Claude Code
Add to your .mcp.json or Claude Code MCP settings:
{
"mcpServers": {
"whatsapp": {
"command": "node",
"args": ["C:/path/to/whatsapp-cli/mcp-server.mjs"]
}
}
}
Run manually
node mcp-server.mjs
CLI Usage
Send a message
node index.js send --to "John Doe" --message "Hey!"
# or by phone number
node index.js send --to 919876543210 --message "Hey!"
Read messages
node index.js read --from "John Doe" --count 20
List chats
node index.js list-chats --count 10
Use Cases
- AI-powered WhatsApp assistant — plug into Claude or any MCP-compatible AI to read and send messages hands-free
- WhatsApp automation — build bots, auto-responders, or notification systems
- Terminal messaging — send/read WhatsApp messages without opening your phone
- Business notifications — send automated WhatsApp alerts from scripts or cron jobs
Project Structure
├── index.js # CLI entry point
├── mcp-server.mjs # MCP server (for AI assistants)
├── src/
│ ├── client.js # WhatsApp client setup
│ └── commands/
│ ├── auth.js # QR code authentication
│ ├── send.js # Send message command
│ ├── read.js # Read messages command
│ └── list-chats.js # List chats command
└── package.json
Requirements
- Node.js 18+
- A WhatsApp account (personal or business)
License
MIT
Related Servers
Fastmail MCP Server
Integrate your Fastmail account with AI assistants using the Model Context Protocol.
MCP IDE Bridge
An open-source messaging server for client-to-client communication using MCP HTTP streaming, configurable via an external JSON file.
Perplexity Chat
An MCP server for the Perplexity API to query responses and manage conversations.
Ntfy
An ntfy MCP server for sending/fetching ntfy notifications to your self-hosted ntfy server from AI Agents 📤 (supports secure token auth & more - use with npx or docker!)
A2A MCP Server
A bridge server connecting Model Context Protocol (MCP) with Agent-to-Agent (A2A) protocol.
MCP Telegram
Telegram MCP server with 20 tools — read chats, search messages, download media via MTProto
Notifications
Send native macOS notifications with full Notification Center integration and interactive features.
Elisym Mcp Server
AI agents that hire other AI agents — and pay in Crypto. Decentralized agent marketplace.
DingTalk
A server for interacting with DingTalk workspaces using the Model Context Protocol.
SendGrid MCP Server by CData
A read-only MCP server by CData that enables LLMs to query live SendGrid data. Requires the external CData JDBC Driver for SendGrid.