快速MCP电报

1. Install and authenticate

uvx --from fast-mcp-telegram fast-mcp-telegram-setup \
  --api-id="your_api_id" \
  --api-hash="your_api_hash" \
  --phone-number="+123456789"

Sessions are stored in ~/.config/fast-mcp-telegram/.

Bot token alternative (no phone number): Create a bot via @BotFather and use --bot-api-token instead of --phone-number:

uvx --from fast-mcp-telegram fast-mcp-telegram-setup \
  --api-id="your_api_id" \
  --api-hash="your_api_hash" \
  --bot-api-token="1234567890:ABCdef..."

No OTP, no verification code. Or skip the setup step entirely — just add BOT_API_TOKEN to your env and the server auto-authenticates on startup.

Deploy to Remote Server

Deploy your own MCP server on a VDS — see Installation Guide for step-by-step instructions.

Quick Start

Try the Demo

1. Open https://tg-mcp.l1979.ru/setup and complete authentication
2. Copy your Bearer token from the setup page

Then choose your path:

MCP Client (AI assistants) - Download the mcp.json file - Add the server to your AI client and ask: "send hello to my saved messages in telegram"

Direct API (curl) - Run the command below (replace TOKEN with yours):

curl -X POST "https://tg-mcp.l1979.ru/mtproto-api/messages.SendMessage" \
  -H "Authorization: Bearer TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"params": {"peer": "me", "message": "Hello!"}}'

2. Configure MCP Client

stdio mode (local):

{
  "mcpServers": {
    "telegram": {
      "command": "uvx",
      "args": ["fast-mcp-telegram"],
      "env": {
        "API_ID": "your_api_id",
        "API_HASH": "your_api_hash"
      }
    }
  }
}

http-auth mode (remote): See Installation Guide for deploying your own server and authenticating via web interface.