WhatsApp Cloud API MCP Server
Provides tools for sending text and template messages, retrieving media files, and checking connectivity with the WhatsApp Cloud API to enable messaging capabilities through Meta's WhatsApp Business platform.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
WhatsApp Cloud API MCP Server – Spec v1.0
Overview
This project defines version 1.0 of an MCP Server that exposes the WhatsApp Cloud API to LLM agents through standardized MCP tools.
Goal: Provide a minimal, clean, production-ready baseline so developers can use WhatsApp messaging inside AI agents with zero complexity.
Related MCP server: MCP WhatsApp
Quick Start
Copy the environment template and fill in your Meta credentials:
cp .env.example .envSet
MCP_AUTH_TOKENto a strong random secret as well - the MCP WebSocket server refuses to start without it.Install dependencies:
npm installRun the server locally:
npm run dev
Tip: if you're testing only the MCP tools locally, use valid WhatsApp Cloud API credentials in
.env. For webhook-only testing, you can use placeholder values while simulating incoming payloads.
1. Objectives
Create a lightweight MCP server (TypeScript + Node.js).
Expose WhatsApp Cloud API functionalities as tools.
Provide helpers for webhook verification.
Be deployable with a single command (Docker + local dev).
2. Tech Stack
Node.js 20+
TypeScript
Express (or Fastify) for HTTP + webhooks
MCP Server SDK
Axios for outbound WhatsApp API requests
WebSocket for MCP transport (mandatory)
Dotenv for configs
Pino or console logging
3. Project Structure (v1.0)
/src
/mcp
server.ts
/whatsapp
client.ts
webhook.ts
/config
env.ts
index.ts
.env.example
mcp.json
package.json
README.md4. Environment Variables
Required:
META_WHATSAPP_TOKEN=
META_WHATSAPP_PHONE_ID=
META_WHATSAPP_WABA_ID=
META_VERIFY_TOKEN=
MCP_AUTH_TOKEN=
PORT=4000
MCP_PORT=8000MCP_AUTH_TOKEN is the shared secret that authenticates MCP WebSocket
clients. The server exits on startup if it is missing.
Optional:
LOG_LEVEL=debug5. WhatsApp Cloud API Endpoints to Support (v1.0)
The MCP server will call these official endpoints:
Send Text Message
POST /v18.0/{PHONE_ID}/messagesSend Template Message
POST /v18.0/{PHONE_ID}/messagesGet Media URL
GET /v18.0/{MEDIA_ID}Download Media
GET {MEDIA_URL}Health Check
GET /v18.0/{PHONE_ID}6. Webhook Handling
Expose:
GET /webhook (verification)Features:
Validate
hub.verify_token
7. MCP Tools (v1.0)
Tools exposed to agents:
1. send_text_message
Send a simple text message to a WhatsApp user.
Params:
{
"to": "string",
"text": "string"
}2. send_template_message
Send a WhatsApp template message.
Params:
{
"to": "string",
"template_name": "string",
"language": "string",
"components": []
}3. get_media
Retrieve a media file by MEDIA_ID.
Params:
{
"media_id": "string"
}4. health_check
Verify connectivity with WhatsApp Cloud API.
8. Security
MCP WebSocket authentication (required). Every WebSocket client must present the shared
MCP_AUTH_TOKEN, either as anAuthorization: Bearer <token>header or as a?token=<token>query parameter (for MCP clients that cannot set headers). Connections without a valid token are rejected during the upgrade handshake, and the server refuses to start whenMCP_AUTH_TOKENis not set. Example:ws://localhost:4000/?token=$MCP_AUTH_TOKENValidate Meta webhook signature (optional v1).
Sanitize user inputs.
Never log tokens.
9. Documentation Requirements
Setup Instructions
Install Dependencies
npm installConfigure Environment Copy
.env.exampleto.envand fill in your Meta WhatsApp API credentials.cp .env.example .envRun Locally
npm run devServer will start on port 4000 (HTTP) and expose MCP on WebSocket.
Configure Webhook
Expose your local server using ngrok or similar:
ngrok http 4000In Meta App Dashboard, set Webhook URL to
https://<your-ngrok-url>/webhookSet Verify Token to match
META_VERIFY_TOKENin.env
Docker Usage
docker build -t whatsapp-mcp .
docker run -p 4000:4000 --env-file .env whatsapp-mcpExample Usage with Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"whatsapp": {
"command": "docker",
"args": ["run", "-i", "--rm", "--env-file", "/path/to/.env", "whatsapp-mcp"]
}
}
}(Note: For local dev without docker, point to the build output)
10. Contact
For any inquiries or similar projects, please contact: marcelo@marcelomarchetti.com
This server cannot be deployed
Maintenance
Related MCP Connectors
WhatsApp for AI agents — your own number or the official Cloud API: messages, media, templates.
- WhatSyncOAuthapp.whatsync
WhatsApp for Claude, ChatGPT or any AI agent: search, reply and send. Personal or Business API.
Read, search and reply to your WhatsApp chats from your AI, using your own number.
WhatsMCP connects Claude and other MCP-compatible AI agents directly to WhatsApp. Send and receive text, images, documents, and voice notes; manage groups (create, add/remove members, promote admins); look up contacts and profiles; follow channels; and read call and message history — all through a standard MCP interface. For voice use cases, WhatsMCP offers SIP-based calling plans (inbound-only, or full inbound/outbound) so AI voice agents can answer and place WhatsApp calls, plus low-latency WebSocket integrations with voice agent providers like ElevenLabs. Multiple WhatsApp accounts can be paired and managed per workspace, with webhook support for real-time inbound message delivery to your own infrastructure.
Related MCP Servers
- AlicenseAqualityFmaintenanceConnects AI assistants to the official Meta WhatsApp Cloud API for managing conversations and sending various message types through natural language. It provides tools for media management, template messages, and real-time webhook processing without the risk of account bans.1815 npm24MIT
- AlicenseBqualityBmaintenanceEnables sending messages, images, documents and more on WhatsApp directly from any MCP-compatible AI, with tools for chat management, groups, and webhooks.371MIT
- AlicenseAqualityDmaintenanceEnables LLMs to interact with WhatsApp via the official WhatsApp Cloud API, providing tools for sending messages, templates, media, and managing conversations.1315 npmMIT
- AlicenseBqualityDmaintenanceEnables AI agents to control WhatsApp, including sending messages and media, reading chats, managing groups and communities, with QR/pairing auth and session persistence.1412 npm2MIT