WhatsApp Claude MCP
Enables WhatsApp bots to process incoming messages through Claude AI, supporting conversational memory and extensible MCP tools for tasks like fetching jokes from external APIs.
README
WhatsApp Claude MCP
A powerful WhatsApp bot integrated with Claude AI using the Model Context Protocol (MCP). Send messages to your WhatsApp bot and get intelligent responses powered by Claude, with access to external APIs and tools.
š Features
- Claude AI Integration: Uses Claude 3.5 Sonnet for intelligent conversations
- MCP Tools: Extensible tool system for Claude to interact with external APIs
- Joke Generator: Built-in tool that fetches random jokes from external API
- Conversation Memory: Maintains context across multiple messages per user
- WhatsApp Webhook: Simple REST API for integrating with WhatsApp services
- Easy Deployment: Works with express server, ready for cloud deployment
š Prerequisites
- Node.js 18+
- npm or yarn
- Anthropic API Key (get from console.anthropic.com)
- WhatsApp Cloud API access (for production integration)
š Quick Start
1. Clone and Install
git clone https://github.com/yulianheroes-lgtm/whatsapp-claude-mcp.git
cd whatsapp-claude-mcp
npm install
2. Set Environment Variables
cp .env.example .env
Edit .env and add your Anthropic API key:
ANTHROPIC_API_KEY=your_anthropic_api_key_here
PORT=3000
3. Start the Server
npm start
You should see:
ā
WhatsApp Claude MCP Server running on http://localhost:3000
š¤ Ready to process WhatsApp messages!
š” API Usage
Health Check
curl http://localhost:3000/health
Send Message to Claude
curl -X POST http://localhost:3000/webhook/whatsapp \
-H "Content-Type: application/json" \
-d '{
"userId": "1234567890",
"message": "Tell me a joke"
}'
Response:
{
"success": true,
"userId": "1234567890",
"message": "š Here's a programming joke for you!\n\nWhy do programmers prefer dark mode?\n\nBecause light attracts bugs! š"
}
Clear Conversation History
curl -X POST http://localhost:3000/webhook/clear-history \
-H "Content-Type: application/json" \
-d '{
"userId": "1234567890"
}'
š ļø Available Tools
Joke Generator
Claude can automatically use this tool when appropriate:
- Trigger: When user asks for jokes
- Types: random, programming, general
- API: Official Joke API
Example interaction:
User: Tell me a funny programming joke
Bot: [Uses joke_generator tool] š Here's a programming joke...
š Project Structure
whatsapp-claude-mcp/
āāā src/
ā āāā index.js # Main Express server
ā āāā whatsapp-handler.js # Message handling & Claude integration
ā āāā mcp-server.js # MCP tool definitions & execution
ā āāā tools/
ā āāā joke-generator.js # Joke generator tool implementation
āāā .env.example # Environment variables template
āāā .gitignore # Git ignore rules
āāā package.json # Dependencies
āāā README.md # This file
š Integration with WhatsApp
Option 1: WhatsApp Cloud API
For production, integrate with WhatsApp Cloud API:
- Set up webhook on Meta Business Platform
- Point webhook URL to:
https://your-domain.com/webhook/whatsapp - When WhatsApp sends messages, forward them to this endpoint
Option 2: Local Testing
Use tools like curl, Postman, or a test script to send messages:
// test.js
const userId = '1234567890';
const message = 'Tell me a joke';
const response = await fetch('http://localhost:3000/webhook/whatsapp', {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({ userId, message })
});
const result = await response.json();
console.log(result.message);
š§ How It Works
- Message Received ā WhatsApp webhook receives a message
- Claude Processing ā Message sent to Claude with available tools
- Tool Selection ā Claude decides if tools are needed
- Tool Execution ā MCP server executes tools (e.g., fetch jokes)
- Response Generation ā Claude generates response using tool results
- Message Sent ā Response sent back via WhatsApp
š Adding More Tools
To add a new tool (e.g., weather, translations):
1. Create Tool File
// src/tools/weather.js
export const weatherTool = {
name: 'get_weather',
description: 'Get current weather for a location',
inputSchema: {
type: 'object',
properties: {
location: { type: 'string', description: 'City name' }
}
}
};
export async function executeWeather(location) {
// Fetch weather data
return { /* weather data */ };
}
2. Register in MCP Server
// src/mcp-server.js
import { weatherTool, executeWeather } from './tools/weather.js';
export class MCPServer {
constructor() {
this.tools = [
jokeGeneratorTool,
weatherTool // Add here
];
}
async processTool(toolName, toolInput) {
switch (toolName) {
case 'get_weather':
return await executeWeather(toolInput.location);
// ...
}
}
}
š API Reference
POST /webhook/whatsapp
Request Body:
{
"userId": "string (required)",
"message": "string (required)"
}
Response:
{
"success": boolean,
"userId": "string",
"message": "string"
}
POST /webhook/clear-history
Request Body:
{
"userId": "string (required)"
}
Response:
{
"success": boolean,
"message": "string"
}
š Security Considerations
- API Keys: Never commit
.envfile to version control - Rate Limiting: Consider adding rate limiting for production
- Input Validation: Always validate webhook payloads
- HTTPS: Use HTTPS in production
- Authentication: Add webhook signature verification for WhatsApp integration
š Environment Variables
| Variable | Description | Example |
|---|---|---|
ANTHROPIC_API_KEY |
Claude API key | sk-ant-... |
PORT |
Server port | 3000 |
NODE_ENV |
Environment | development |
JOKE_API_URL |
Joke API endpoint | https://official-joke-api.appspot.com/random_joke |
š¤ Contributing
Feel free to fork, modify, and contribute back!
š License
MIT License - see LICENSE file for details
š Troubleshooting
"API key not found"
- Ensure
.envfile exists and hasANTHROPIC_API_KEYset - Check key is valid at console.anthropic.com
"Tool execution failed"
- Check external APIs are accessible
- Verify network connectivity
- Review error logs in console output
"No response from Claude"
- Check
ANTHROPIC_API_KEYis correct - Ensure Claude model is available
- Check API rate limits
š Support
For issues or questions:
- Check the troubleshooting section
- Review Claude API documentation
- Open an issue on GitHub
šÆ Future Enhancements
- [ ] Support for image/media in WhatsApp messages
- [ ] Additional tools (weather, news, translations)
- [ ] Database for persistent conversation history
- [ ] Rate limiting and authentication
- [ ] Admin dashboard for monitoring
- [ ] Multi-language support
- [ ] Custom Claude system prompts per user
Made with ā¤ļø by yulianheroes-lgtm
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.