gaviwhatsapp-mcp
Send WhatsApp messages, manage templates, and run broadcast campaigns from AI coding tools. Works with Cursor, Claude Code, and Codex via npx @gaviwhatsapp/mcp.
README
Gavi WhatsApp API
The WhatsApp API built for how developers actually build in 2026. Works in Cursor, Claude Code, Codex, and Lovable.
MCP server · TypeScript SDK · Python SDK · Examples · $9.99/mo
What is this?
A complete WhatsApp Business API platform with:
- MCP Server — Type "send a WhatsApp message" in Cursor or Claude Code and it just works
- REST API — 6 endpoints: send message, send template, send media, broadcasts, webhooks, contacts
- TypeScript SDK —
npm install @gaviwhatsapp/whatsapp - Python SDK —
pip install gaviwhatsapp - Dashboard — Visual form builder, conversation flows, broadcast campaigns, analytics
Quick Start
Option 1: MCP Server (AI coding tools)
Add to .cursor/mcp.json or .claude/mcp.json:
{
"mcpServers": {
"gaviwhatsapp": {
"command": "npx",
"args": ["@gaviwhatsapp/mcp", "--api-key", "gv_YOUR_KEY"]
}
}
}
Then just ask: "Send a WhatsApp message to +919876543210 saying hello"
Option 2: TypeScript SDK
npm install @gaviwhatsapp/whatsapp
import { WhatsApp } from '@gaviwhatsapp/whatsapp'
const wa = new WhatsApp({ apiKey: process.env.GAVIWHATSAPP_API_KEY! })
await wa.send({ to: '+919876543210', text: 'Hello from my app!' })
Option 3: Python SDK
pip install gaviwhatsapp
from gaviwhatsapp import WhatsApp
wa = WhatsApp(api_key="gv_YOUR_KEY")
wa.send(to="+919876543210", text="Hello from my app!")
Option 4: REST API
curl -X POST https://www.gaviventures.com/api/v1/messages \
-H "Authorization: Bearer gv_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{"to": "+919876543210", "text": "Hello!"}'
Get an API Key
- Sign up at gaviventures.com
- Connect your WhatsApp Business Account
- Go to Settings > API Keys
Packages
| Package | Description | Install |
|---|---|---|
@gaviwhatsapp/mcp |
MCP server for Cursor, Claude Code, Codex | npx @gaviwhatsapp/mcp |
@gaviwhatsapp/whatsapp |
TypeScript/Node.js SDK | npm install @gaviwhatsapp/whatsapp |
gaviwhatsapp |
Python SDK | pip install gaviwhatsapp |
Examples
| Example | Description | Language |
|---|---|---|
| send-message-ts | Send a text message | TypeScript |
| send-message-python | Send a text message | Python |
| broadcast-campaign | Send personalized broadcast from CSV | TypeScript |
| webhook-receiver | Receive incoming WhatsApp messages | TypeScript |
API Reference
| Method | Endpoint | Description |
|---|---|---|
| POST | /api/v1/messages |
Send a text message |
| POST | /api/v1/messages/template |
Send a template message |
| POST | /api/v1/messages/media |
Send media (image/video/document) |
| POST | /api/v1/broadcasts |
Send to multiple recipients |
| POST | /api/v1/webhooks |
Register a webhook |
| GET | /api/v1/webhooks |
List webhooks |
Full docs: gaviventures.com/docs/api
Pricing
| Plan | Price | Includes |
|---|---|---|
| API | $9.99/mo | API, SDKs, MCP server, webhooks, broadcasts |
| Pro | $24.99/mo | Everything + visual builder, forms, flows, analytics |
7-day free trial. No credit card required. Meta per-message charges billed directly to your WhatsApp Business Account.
Links
License
Source code in this repository is provided for reference and integration purposes. The WhatsApp API service requires a paid subscription. See gaviventures.com for terms.
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。