telegram-mcp
A minimal Model Context Protocol server for interacting with Telegram bots via MTProto. It provides simple tools to send messages and retrieve chat history while serving as an easy-to-extend reference for developers.
README
telegram-mcp
A minimal MCP server for Telegram bot interaction via MTProto — intentionally simple, easy to read, and easy to extend.
Looking for a full-featured Telegram MCP server? Check out chigwell/telegram-mcp — 60+ tools covering messaging, groups, contacts, media, admin, and more.
This project takes a different approach: two tools, ~150 lines of server code, zero complexity. It's a clean starting point for anyone who wants to understand how MCP servers work with Telegram, or who only needs basic bot messaging without the overhead of a full client.
Why this exists
- Learning reference — Read the entire server in 5 minutes
- Fork-friendly — Add exactly the tools you need, nothing you don't
- Minimal dependencies — Just Telethon, MCP SDK, and python-dotenv
Tools
| Tool | Description |
|---|---|
send_message |
Send a message to a Telegram bot and wait for its reply |
get_history |
Get message history with a Telegram bot |
Setup
1. Get Telegram API credentials
- Go to https://my.telegram.org/apps
- Log in with your phone number
- Create a new application (any name/description)
- Copy the API ID and API Hash
2. Install and authenticate
# Install
pip install git+https://github.com/tensakulabs/telegram-mcp.git
# Set your credentials
export TELEGRAM_API_ID=12345678
export TELEGRAM_API_HASH=abcdef1234567890abcdef1234567890
export TELEGRAM_PHONE=+15551234567
# Run one-time authentication (sends a code to your Telegram app)
telegram-mcp-auth
3. Configure your MCP client
Claude Code
Add to ~/.claude/settings.json:
{
"mcpServers": {
"telegram": {
"command": "uvx",
"args": ["--from", "git+https://github.com/tensakulabs/telegram-mcp.git", "telegram-mcp"],
"env": {
"TELEGRAM_API_ID": "your-api-id",
"TELEGRAM_API_HASH": "your-api-hash",
"TELEGRAM_SESSION_DIR": "/path/to/session/dir"
}
}
}
}
Claude Desktop
Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"telegram": {
"command": "uvx",
"args": ["--from", "git+https://github.com/tensakulabs/telegram-mcp.git", "telegram-mcp"],
"env": {
"TELEGRAM_API_ID": "your-api-id",
"TELEGRAM_API_HASH": "your-api-hash",
"TELEGRAM_SESSION_DIR": "/path/to/session/dir"
}
}
}
}
Local development
If you've cloned the repo:
{
"mcpServers": {
"telegram": {
"command": "uv",
"args": ["--directory", "/path/to/telegram-mcp", "run", "telegram-mcp"],
"env": {
"TELEGRAM_API_ID": "your-api-id",
"TELEGRAM_API_HASH": "your-api-hash"
}
}
}
}
Configuration
| Environment variable | Required | Default | Description |
|---|---|---|---|
TELEGRAM_API_ID |
Yes | — | Your Telegram API ID |
TELEGRAM_API_HASH |
Yes | — | Your Telegram API hash |
TELEGRAM_PHONE |
Auth only | — | Phone number for authentication |
TELEGRAM_ENV_PATH |
No | Auto-detect | Path to .env file |
TELEGRAM_SESSION_DIR |
No | Working directory | Directory for session file |
How it works
This server uses the Model Context Protocol to expose Telegram interaction as tools that any MCP-compatible AI client can use. Under the hood, it:
- Connects to Telegram via MTProto using your pre-authenticated session
- Sends messages to bots as your user account
- Polls for replies and returns structured JSON responses
- Disconnects cleanly after each operation
Security notes
- Your session file (
telegram_user.session) contains auth tokens — never commit it - The
.gitignoreexcludes session files by default - API credentials should be passed via environment variables, not hardcoded
- This tool acts as YOUR Telegram account — treat it accordingly
Development
git clone https://github.com/tensakulabs/telegram-mcp.git
cd telegram-mcp
uv sync
uv run telegram-mcp
Requirements
- Python 3.10+
- uv (recommended) or pip
License
MIT
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.
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.
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.
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.