Telegram Notifier MCP Server
Enables AI agents to send notifications and media (text, photos, documents, videos) via a Telegram bot.
README
Telegram Notifier MCP Server
A Model Context Protocol (MCP) server that enables AI agents to send messages via Telegram bot. Perfect for notifications, alerts, and real-time communication from your AI applications.
Features
- 📱 Send messages to Telegram channels/chats via bot
- 🎨 Support for multiple formatting modes (Markdown, MarkdownV2, HTML)
- 📷 Send photos/images (local files, HTTP URLs)
- 📁 Send documents/files with custom filenames
- 🎬 Send videos with caption support
- 🏷️ Caption support for all media with full formatting
- 🛡️ Robust error handling with detailed Telegram API responses
- 🏗️ Clean service architecture for maintainable code
Quick Start
1. Create a Telegram Bot
- Message @BotFather on Telegram
- Use
/newbotcommand and follow instructions - Save the bot token (looks like
123456789:ABCdefGHIjklMNOpqrsTUVwxyz)
2. Get Chat ID
For personal messages:
- Message your bot first
- Visit:
https://api.telegram.org/bot<YOUR_BOT_TOKEN>/getUpdates - Find the
chat.idvalue
For channels:
- Add your bot to the channel as admin
- Send a message to the channel
- Visit the same URL and find the
chat.id(negative number for channels)
3. Install and Configure
# Install the package
npm install -g @harnyk/telegram-notifier-mcp
# Or clone and build locally
git clone <repository-url>
cd mcp-telegram-notifier
pnpm install
pnpm build
pnpm link
4. Add to Your MCP Client
Add to your MCP client configuration (e.g., Claude Desktop):
{
"mcpServers": {
"telegram-notifier": {
"command": "telegram-notifier",
"env": {
"TELEGRAM_BOT_TOKEN": "your_bot_token_here",
"TELEGRAM_CHAT_ID": "your_chat_id_here"
}
}
}
}
Or using npx:
{
"mcpServers": {
"telegram-notifier": {
"command": "npx",
"args": ["-y", "@harnyk/telegram-notifier-mcp"],
"env": {
"TELEGRAM_BOT_TOKEN": "your_bot_token_here",
"TELEGRAM_CHAT_ID": "your_chat_id_here"
}
}
}
}
Available Tools
Once configured, your AI agent will have access to these MCP tools:
📝 send_markdown_message_as_telegram_bot
Send formatted text messages to Telegram
- messageText: Text content with Markdown/HTML formatting
- parseMode:
Markdown,MarkdownV2, orHTML(default: MarkdownV2)
📷 send_telegram_photo
Send photos/images to Telegram
- photo: Local file path or HTTP URL to image
- caption: Optional image caption with formatting
- parseMode: Caption formatting mode
📁 send_telegram_document
Send documents/files to Telegram
- document: Local file path or HTTP URL to document
- caption: Optional document caption with formatting
- filename: Custom filename override
- parseMode: Caption formatting mode
🎬 send_telegram_video
Send videos to Telegram
- video: Local file path or HTTP URL to video
- caption: Optional video caption with formatting
- filename: Custom filename override
- parseMode: Caption formatting mode
Use Cases
- 📊 Monitoring: Get alerts when system metrics exceed thresholds
- 🚀 Deployments: Notifications when CI/CD pipelines complete
- 📈 Analytics: Daily/weekly reports with charts and data files
- 🔍 Error Tracking: Instant notifications with error screenshots
- 📅 Scheduling: Reminders with attached documents
- 💬 Customer Support: Forward important messages with media
Development
# Install dependencies
pnpm install
# Run in development mode
pnpm dev
# Build for production
pnpm build
# Run linting
pnpm lint
# Run type checking
pnpm typecheck
Configuration Examples
Claude Desktop
{
"mcpServers": {
"telegram-notifier": {
"command": "telegram-notifier",
"env": {
"TELEGRAM_BOT_TOKEN": "123456789:ABCdefGHIjklMNOpqrsTUVwxyz",
"TELEGRAM_CHAT_ID": "-1001234567890"
}
}
}
}
Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests if needed
- Run
pnpm lintandpnpm typecheck - Submit a pull request
License
WTFPL - Do What the Fuck You Want to Public License
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。