The terms under which you may use mctl-telegram. See /privacy for data handling and /security for the threat model.
mctl-telegram is an open-source MCP server that lets an AI assistant act on your behalf against your own Telegram user account over MTProto. It is an independent project. It is not affiliated with, endorsed by, or operated by Telegram or OpenAI, and "Telegram" and "ChatGPT" are trademarks of their respective owners.
On the hosted deployment at tg.mctl.ai, sending is gated and dry-run by default: send_message returns a preview (sent=false) instead of delivering anything. Real delivery happens only when every opt-in gate is enabled (server ALLOW_SEND, your account send_enabled flag, and the telegram:messages:send scope). Once those are enabled, the service will send real messages to real people on your behalf, and a sent message cannot be unsent. Enable sending only if you accept that.
The service is provided "as is" and "as available", without warranties of any kind, express or implied, including fitness for a particular purpose and non-infringement. Parts of it (including Local Bridge mode) are beta and may change or break. To the maximum extent permitted by law, the operator and contributors are not liable for any loss or damage arising from your use of the service, including account limitations, lost or misdirected messages, or downtime.
There is no uptime guarantee. The service may be modified, suspended, or discontinued at any time. These terms may be updated; continued use after a change means you accept the updated terms. You can stop using the service and remove your data at any time via disconnect_telegram_account / delete_telegram_account or the HTTP endpoints listed on /privacy.
General support: [email protected]. For security or vulnerability reports, see /security. The source is open at github.com/mctlhq/mctl-telegram — audit it, self-host it, or contribute.