Telegram Bot Integration

Connect your Telegram bot to a project, then use an active Telegram Flow to define how it answers messages.

An active Telegram Flow is required

Connecting a bot enables the Telegram channel. Its replies, knowledge-base search, AI calls, and conversation steps are configured in Flows. If the project has no active Telegram Flow, incoming messages are ignored and the bot does not reply.

Before you start

  • A FileBrain.Pro project and a Telegram account.
  • A bot token from @BotFather.
  • Project knowledge-base materials if your Flow will use RAG Search, and enough credits for the Flow's operations.

Connect the bot and activate its Flow

  1. Create a Telegram bot

    Open @BotFather in Telegram, send /newbot, and follow its instructions to choose a name and username. Copy the bot token it provides.

  2. Connect it to your project

    Open your project's Integrations page and choose Telegram Bot. Paste the token and save the connection. FileBrain.Pro configures message delivery automatically.

  3. Create a Telegram Flow

    In the same project, open Flows, create a new Flow, and choose the Telegram trigger. Each project can have one active Telegram Flow.

  4. Build and activate the conversation

    Start with Telegram Input → Telegram Response. Connect the nodes and enter a short test reply in the response node's Response Text field. Save the Flow and make it active.

    For AI answers, add the processing nodes your bot needs, such as RAG Search or Invoke LLM, and use the variable picker to pass their result to Telegram Response.

  5. Test in Telegram

    Open your bot's chat and send a text message. Check that the expected reply arrives, then review the project's Flow execution logs. Test the complete conversation before sharing the bot with users.

See Getting Started with Flows for the editor basics and the Telegram Flow connector guide for available nodes and conversation behavior.

Add the behavior your bot needs

  • Telegram Input makes the message text, session, chat, and user details available to the Flow.
  • RAG Search and Invoke LLM let you build knowledge-base and AI answers.
  • User Input asks a question and pauses the Flow until the next user reply.
  • Telegram Response returns the final answer. Telegram Message can send an additional message during processing.
  • Add Read Memory and Append Memory nodes if the Flow needs conversation history.

Already connected a bot?

Keep the existing connection and create or activate a Telegram Flow in that project. The former direct-answer mode is no longer available. Connecting the bot or uploading documents alone does not configure its replies.

If the bot does not answer

  • Check that the bot is connected to the same project as the Flow.
  • Check that the Flow uses the Telegram trigger, is saved, and is active.
  • Check the node connections and the reply configured in Telegram Response.
  • Review the Flow execution logs and available credits. Incoming text messages are supported; attachments are not processed by this connector.

For further help, contact support.