Telegram auto-reply without Premium: connect it in 8 steps

What an AI auto-reply in Telegram is

An auto-reply helps handle incoming messages without a person being involved every time. The simplest kind sends one pre-written sentence:

That is useful as a notification, but it does not understand the question and does not continue the conversation. An AI auto-reply works differently: it reads what the message means, takes the conversation into account, picks a fitting profile and answers from the knowledge you added. “How much is a consultation?” gets your confirmed terms and an offer of the next step; “I cannot log in to my account” gets a clarifying question and a request for a screenshot; “We would like to propose a partnership” gets the key details collected for the owner.

The kinds of auto-reply you can use

Templates save time when you often send the same information, but a person still picks and sends the message. An ordinary bot copes well with menus and step-by-step flows, though a freely worded question may not match any command it was given. An AI assistant reads what the message means, carries on with the context in mind and uses the knowledge you added.

If you are comparing this with Telegram's own features — the greeting and the “Away” status — a full walkthrough of all three routes is in the article “How to set up a Telegram auto-reply without programming”.

When an AI auto-reply is genuinely worth it

The assistant earns its place when people word the same question in different ways and the answer depends on context. It also helps when a conversation needs your price list, instructions, facts and different scenarios for clients, support or personal contacts.

If the job comes down to occasionally sending an address or payment details, templates are enough. If a person is meant to press buttons in order and fill in a simple form, an ordinary Telegram bot fits. An AI auto-reply is justified where free text has to be understood, the right knowledge chosen and the conversation carried on without a rigid menu.

Can HeadHack be used without Telegram Premium

Yes. HeadHack does not require a Telegram Premium subscription. The setup runs through the ready-made @Head_Hack_bot and the mini app inside Telegram.

You also do not have to write code, create a separate bot through Telegram's own bot service, run a server or wire up a model API yourself.

What Telegram Premium gives and what works without it

The subscription gates one specific route — Telegram's own business features — rather than auto-replies as such. Everything the bot does is independent of it.

The difference fits in one line: without Premium you lose Telegram's pre-written texts, not the auto-replies themselves. A meaningful answer to a client's question is the assistant's work either way, and it does not depend on a subscription.

An auto-reply without Premium works where it works with it: in direct conversations. The assistant answers new incoming direct text messages in the chats you allowed; in groups, channels and comments it does not. Voice notes it transcribes and answers as ordinary text.

The device makes no difference: the setup happens in the mini app and the replies are prepared in the cloud, so the auto-reply works with your phone switched off. Step-by-step menu paths for Android, iPhone and desktop are in the guide “Telegram auto-reply: 3 ways”.

Why plugging in GPT is not enough on its own

It can look as though an auto-reply needs one instruction: “answer every incoming message”. In practice such a setup goes wrong fast.

“How much is it?” may be about a consultation, a subscription, delivery, development work or one particular plan. “It stopped working after the update” does not say what broke either: the app, the website, the account area, payments or one specific feature.

Hand the model everything at once under a single general instruction and the familiar problems appear: the assistant picks information from the wrong topic, a client gets an answer meant for a partner or a candidate, a personal message comes out stiffly formal, missing data gets filled in with invention, and the model cannot tell when to ask a question and when to hand the conversation to a person.

Profiles solve this: each kind of message gets its own topics, knowledge, documents, limits and tone.

How HeadHack works: profiles

One blanket instruction to “answer everything” is usually not enough. The same question can belong to different topics: a client asks about price, a user reports a bug, a partner proposes a deal, a relative writes about something personal.

That is why HeadHack uses profiles. Each profile holds its own rules and knowledge for one kind of conversation.

For a question about price the system may choose the “Clients” profile, and for a bug report the “Support” one. More on splitting clients, support, partners and personal messages is in the article “Profiles in HeadHack”.

A good auto-reply runs the whole route: it reads the incoming message, takes the conversation into account, picks a profile, applies the system prompt, uses the FAQ, facts and documents, answers in the chosen style and offers a next step — or hands the question to a person.

What you can set in a profile

The whole list only needs filling in once per profile. There is no need to repeat the same details in several sections.

Which profile to start with

You do not have to move every kind of conversation into HeadHack at once. Pick one scenario where the questions repeat most often and the answer is easy to check. Usually that is the profile for clients or for support.

For a first profile it is enough to prepare:

• a clear name;
• five or six specific topics;
• a short description of the reply style;
• one system prompt with no contradictions;
• a handful of common questions;
• the key facts;
• one or two current documents;
• a rule for when the information runs out.

Once the first profile runs steadily, add the next. That order makes it easy to see where a mistake came from: the topic, the prompt, a fact or a document.

How to connect a Telegram auto-reply: 8 steps

1. Open HeadHack. Go to @Head_Hack_bot and press Start.
2. Open the mini app. Every important setting lives inside it.
3. Create a profile. Give it a clear name that says what kind of conversations it covers: “Clients”, “Support”, “Partners”.
4. Set the rules. Describe the reply style and write the system prompt.
5. Add knowledge. Fill in the FAQ, the facts and the profile's topics.
6. Attach documents. Upload .pdf, .docx, .txt or add a public link.
7. Switch the profile on.
8. Test the replies on real examples, and connect Google Calendar if you need bookings.

Before going fully live it is worth checking how the profile answers an exact question, an incomplete phrase and a request that is not in the knowledge base at all. A detailed checklist is in the article “How to test an AI auto-reply before launch”.

What the bot can do

1. answer questions about your services and terms;
2. collect requests and clarify what is needed;
3. help book a consultation or a meeting;
4. send instructions, documents and links;
5. handle support enquiries;
6. hand a conversation to a manager, a specialist or you;
7. follow personal scenarios for relatives and important contacts.

The main job is not simply to hand over information but to help the person take one clear next step. Detailed scenarios for sales, requests and bookings are in the article “A Telegram bot for requests, sales and bookings”.

How to switch the auto-reply off without Premium

Switching off takes one movement and erases nothing: profiles, knowledge and documents stay where they are, the assistant simply stops answering.

• Everywhere. Turn off the assistant's toggle in the mini app. Replies stop in every chat at once.
• In one conversation. Remove the chat from the allowed list. The auto-reply keeps working in the rest.
• For one kind of enquiry. Switch the profile off. Keep “Clients” and disable “Support” while you are away, for instance.

If you were using Telegram's own auto-replies before, they have to be switched off separately — they live in the Telegram Business settings and have nothing to do with HeadHack. The steps are in the section “How to remove or switch off a Telegram auto-reply”.

The auto-reply is silent: what to check

1. The assistant is switched off. Check the toggle in the mini app — this is the most common reason.
2. The chat is not on the allowed list. The assistant only answers where you allowed it. It does not read the other conversations at all.
3. You are writing to yourself. Your own messages get no reply. Test from another account.
4. The message is not a direct one. A group, a channel or a comment under a post is not the case here: the auto-reply works in direct conversations.
5. The plan's messages ran out. The free plan has 50 a month. Once the limit is reached the assistant goes quiet until next month or until you move to a paid plan.

If the assistant answers but answers wrongly, that is not a broken connection but a profile setting: check which profile the conversation falls under and whether the fact you need is in the knowledge base.

What to check before launch, and about safety

Before switching a profile on, check:

• whether the reply style is clearly written;
• whether the system prompt contradicts itself;
• whether the FAQ and facts are current;
• whether topics and roles really separate one profile from another;
• whether the documents are cluttered with outdated versions;
• whether the calendar is set up correctly;
• whether the assistant asks one clarifying question when data is missing;
• whether every reply leads to a clear action.

Telegram auto-reply examples: ready-made texts

Setting the auto-reply up is half the job: next you have to decide what it will write. Below are eight replies to the most common situations, shown as they look in a conversation. Copy them as they are or rewrite them in your own voice — the HeadHack assistant answers in your style and fills in the details from your knowledge base.

A good auto-reply does three things: it takes the worry away (“you have been heard”), it names a time, and it offers a next step. A reply with neither a time nor a step reads as a brush-off.

Common questions

Related reading

Detailed walkthroughs of the individual HeadHack settings live in seven articles in this series:

What to choose in the end

Saved templates suit rare, identical replies. An ordinary Telegram bot suits menus, forms and simple requests. An AI auto-reply is what you need when free text has to be understood, context taken into account and the conversation steered towards an action.

Open @Head_Hack_bot, create your first profile, add a style, a prompt and some knowledge, then try it on a few real conversations.

What the assistant writes in your name, how it answers the direct question “is this a bot?” and how to read every reply before it is sent — in the article “How to keep a client from realising a bot is answering”.