Telegram MCP is a server tool by Chaindead, built with Node.js, that connects AI models and clients to the Telegram messaging platform. Categorized as an MCP server, it acts as a crucial bridge, enabling AI to interact with the Telegram API. It is available through binary releases for macOS, Linux, and Windows, installable via Homebrew, runnable with NPX, or by cloning its GitHub repository. This tool empowers AI agents to extend their reach into Telegram chats and automate communication workflows.

What does Telegram MCP connect to?

The Telegram MCP server connects directly to the Telegram messaging platform. It functions as an MCP (Model Context Protocol) server, specifically designed to bridge AI models and clients with the Telegram API. This connection allows various AI applications to send, receive, and manage messages within Telegram, integrating AI capabilities into a widely used communication channel. It establishes a necessary link for automated interactions and data exchange with the platform.

Core features and tools

Telegram MCP exposes several callable tools for interacting with Telegram. These include tg_me for current account information, tg_dialogs to list conversations with an optional unread filter, and tg_read to mark a conversation as read. It also offers tg_dialog for retrieving messages from a specific conversation and tg_send to dispatch draft messages. Additionally, the server supports executeMethod for running arbitrary Telegram API methods, expanding its utility for developers.

Who is it for?

Telegram MCP is for AI agents and other MCP clients, particularly users and development teams aiming to connect AI models to Telegram chats. It serves those who wish to automate communication workflows within the Telegram ecosystem. Compatible AI clients include Cursor, Claude Desktop (Claude Cowork or Claude Work), ChatGPT, Augment Code, and Cline. This is not the right fit if your goal is to interact with Telegram without involving AI models or other MCP clients.

How to get started

The server authenticates using the user's Telegram phone number and a confirmation code, with support for two-factor authentication. Communication occurs via stdio when configured by its command, or HTTP for multiple MCP clients.

Use cases

• Sending welcome messages to new group members.
• Summarizing unread Telegram messages and drafting responses.
• Automating release communications or team coordination tasks.

Telegram MCP provides a direct pathway for AI models and clients to engage with the Telegram platform. Its specific tools and authentication model enable automated interactions and communication workflows. This server tool is a valuable asset for integrating AI into messaging tasks. Start by exploring its capability to summarize unread Telegram messages and draft responses.