All features

AI conversation portal: creator replies on one screen, handover is yours

When creators reply, the conversation lands in the panel; if the agency turns the assistant on, a reply based on your knowledge base is prepared, and the moment you write, the conversation is yours.

  • The Messages screen and replies

    Outgoing messages, replies and the chats the assistant hands to you, on one screen.

Five steps from a reply to a handover

  1. Step 1:

    The reply lands in the panel

    The creator's reply is matched to the creator's record on the Messages screen. The earlier messages and who the creator is sit on the same screen.

  2. Step 2:

    The message is classified

    If the assistant is on, an incoming message is classified first: a question, small talk, no interest or unclear. The model cannot compose sentences; it picks only one label from a closed list.

  3. Step 3:

    A reply is prepared

    Which label leads to which outcome is written in a fixed table in the system. The reply is based on the knowledge base your agency approved; for some questions an approved answer is used as it is.

  4. Step 4:

    It passes the rules

    A prepared reply is not used until it passes fixed rules that contain no AI. A reply that fails a rule does not go out, and the conversation is left to your team.

  5. Step 5:

    You take over

    The moment you write in a conversation from the panel, it switches to human mode and any waiting assistant reply is cancelled. A message the system is unsure about is marked "waiting for a human".

What you get

Replies do not wait for morning

Creators often come back after a stream, late at night or at the weekend. With the assistant on, frequent questions do not wait for the morning, and your team looks at the conversations that need a decision.

You stay in control

Every conversation has a mode: assistant, human or silent. You can leave the assistant off entirely or switch single conversations to human mode; the panel is then the shared screen where replies collect.

The assistant has no authority

The model only suggests text. It cannot change records, turn on settings, send a message itself or invent a link. The system adds the application link from the agency's record.

Three conversation modes: assistant, human, silent

  • Assistant

    A reply that passed the rules is used

  • Human

    Your team writes the answer

  • Muted

    No reply is prepared

Read more

In assistant mode, a reply that passed the rules is used. Model replies also depend on a switch the agency turns on and off; with the switch off no model reply goes out, while messages your team writes from the panel keep going.

In human mode the assistant does not write to that conversation, and your team writes the answers. The moment an operator writes in a conversation it switches to human mode; the team and the model never write over each other in the same conversation. In silent mode no reply is prepared: a conversation with harassment, abuse, spam or a scam attempt is muted. A creator who asks you not to write is added to your block list.

The details

Knowledge base

The assistant relies on the agency's knowledge base: what the agency does, how earnings work, gifts, conditions, stream hours and how to apply. Approved answers added to the knowledge base pass the same rules when they are saved.

Read more

When an application link is needed, the model does not write the address; it only signals that a link is needed, and the system adds the address from the agency's record. A reply that carries a link or an address is rejected by the rule layer.

Questions about this feature

Does the assistant send invitations?

No. TT Panel does not send invitations. The assistant only prepares replies to questions in a conversation; it has no authority to change records or turn on settings in the panel.

Can I keep the assistant off?

Yes. The assistant is a layer the agency turns on and off. With it off, no model reply goes out; the panel is the shared screen where replies collect and your team writes the answers.

What if the assistant prepares a wrong reply?

Replies rest on the knowledge base your agency approved and pass fixed rules before sending. A reply that fails a rule does not go, and the conversation is left to your team. The whole conversation stays in the panel, so you can take over any time.

How do I take over a conversation?

Just write in that conversation from the panel. The moment you write, it switches to human mode and the waiting assistant reply is not used. If you want to hand it back to the assistant, change the mode.

Does the assistant write the first message too?

No. When the first message goes out from a template, your agency decides the text; if you prefer, your team writes it. The assistant only helps prepare replies in conversations where the creator has answered, and only if the layer is on.

Contact us with your questions or for detailed information about the system

Write to us on WhatsApp or by email; we answer your questions and explain the system in your own language.