🇬🇧 English · 🇷🇺 Русский

Telegram MCP

Catch up on a conversation and prepare your reply in the same chat.

Find an earlier agreement, collect context from a busy conversation or review incoming messages before replying. Telegram MCP connects your assistant to a Telegram user account. It starts with read-only access; sending can be enabled with separate permissions.

Quick start · Connect your assistant · Issues

Try asking your assistant:

Read the recent messages in the chat I select. Summarise the agreements and open questions, then draft a reply for me to review.

What you can do

Related MCP server: agent-telegram-mcp

Quick start

Prefer a ready package? Install the release and generate your client configuration. No source checkout is required.

Install Python 3.12–3.14 and uv. Clone with Git or download the ZIP. With a ZIP, open the extracted directory and skip the first two commands.

You need a Telegram API ID, API hash and an authenticated user session. The local wizard can create the session through an interactive login. A bot token cannot replace the user session. See Telegram login setup.

git clone https://github.com/zai-one/telegram-mcp.git
cd telegram-mcp
uv sync --frozen
uv run --frozen python scripts/configure.py
uv run --frozen zai-telegram-mcp --config mcp.local.json --check-config

The wizard creates a local configuration and stores secrets in private files. It refuses to overwrite an existing setup. --check-config validates local settings; the first request below checks your account connection.

Connect your assistant

Add this configuration to an MCP client that uses mcpServers, such as Claude Desktop or Cursor. Replace /ABSOLUTE/PATH/ with your absolute path; Windows JSON paths can use forward slashes, such as D:/Tools/.

{
  "mcpServers": {
    "telegram": {
      "command": "uv",
      "args": [
        "--directory",
        "/ABSOLUTE/PATH/telegram-mcp",
        "run",
        "--frozen",
        "zai-telegram-mcp",
        "--config",
        "/ABSOLUTE/PATH/telegram-mcp/mcp.local.json"
      ]
    }
  }
}

The client starts the MCP server for you. Refresh its tool list, then make your first request. For clients with a different config format, reuse the same command and args; uv must be available to the client process.

First request

Show my recent Telegram chats. Ask which one to open before reading its messages.

A valid user session returns a list of chats. Choose one to retrieve messages or conversation context. Drafting and summarising happen in your AI assistant; the MCP server supplies Telegram data and supported actions.

If tools do not appear, check the absolute path, whether the client can find uv, and the --check-config result. For access errors, check account credentials and permissions. Installation and troubleshooting.

Access and limits

Your session grants access to Telegram data, so use a client you trust. The assistant receives messages you request. Incomplete recipient searches remain ambiguous. Media setup and message polling explain file permissions, size limits and acknowledgment. Polling tracks new message IDs; it does not track edits or deletions and is not a live event stream.

Authenticated HTTP is available for a server deployment. See HTTP setup, configuration and permissions and Python package integration.

uv sync --frozen --all-groups
uv run --frozen python scripts/verify.py
uv run --frozen python scripts/verify_install.py

Tests use synthetic fixtures. A passing test run does not establish live provider connectivity.

Built by ZAI.ONE

ZAI.ONE is a digital agency working on websites, SEO, advertising and analytics. We also build tools that connect AI assistants to everyday work. Talk to us on Telegram about setup, automation or an integration for your team.

Use and feedback

You may install and use this project for your own accounts under LicenseRef-ZAI-ONE. This is not an open-source license. Third-party notices remain in NOTICE. If it helps, give the repository a ⭐. Missing something or found a bug? Open an issue. I'm working on this project; accepted improvements are implemented here. Support is not guaranteed.