Flaex Logo
Flaex.ai
Back to MCP Directory
Telegram
Model Context Protocol moderate risk

Telegram

This MCP server bridges AI assistants with the Telegram API, enabling message management, organization, and communication, but requires careful API key management.

Connections & Capabilities

Connects To

GitHub

Capabilities

readwrite

Quickstart

Install

npx -y @chaindead/telegram-mcp

Config

{
  "mcpServers": {
    "telegram": {
      "command": "npx",
      "args": ["-y", "@chaindead/telegram-mcp"],
      "env": {
        "TG_APP_ID": "<your-api-id>",
        "TG_API_HASH": "<your-api-hash>"
      }
    }
  }
}

Exposed MCP Tools (5)

safe
tg_me

Retrieves information about the current Telegram account.

Read-only operation, no side effects.

safe
tg_dialogs

Lists Telegram dialogs, optionally filtering by unread status.

Read-only operation, no side effects.

moderate
tg_read

Marks a specific Telegram dialog as read.

Modifies dialog state, but not destructive.

safe
tg_dialog

Retrieves messages from a specific Telegram dialog.

Read-only operation, no side effects.

moderate
tg_send

Sends a draft message to a specified Telegram dialog.

Sends messages; potential for unintended communication.

Safety Assessment

This server offers useful Telegram integration, but the ability to send messages introduces risk. It's relatively safe for read-only operations, but sending messages requires careful prompt engineering and monitoring to prevent unintended actions. Secure storage of API keys is crucial.

  • Requires explicit API ID and hash for authorization.
  • Provides tools for message reading and sending, but not deletion.
  • User must grant permissions via Telegram's authentication flow.
  • Clear separation of concerns via MCP.
  • Improper handling of API credentials can lead to account compromise.
  • Sending messages through the API can have unintended consequences if not carefully controlled.
  • No built-in rate limiting; potential for abuse of Telegram API.
  • Server relies on external Telegram API, introducing a dependency risk.
  • Lack of sandboxing could expose the host system to vulnerabilities if the server code is compromised.