Lovable chatbot for support: build it or add one

Lovable doesn’t come with a chatbot for your users, but it will build you one. Ask in the project chat and Lovable writes the chat window, the backend function that calls the model and the tables that hold your docs, and its built-in AI connector runs the model with no API key to manage. Every reply that chatbot writes is paid from your Lovable credits, the same balance you build with, and it stops answering when that balance reaches zero. We read Lovable’s docs, guides and pricing on 23 September 2026.

Deacon is an AI customer support agent you add to a Lovable app by pasting one line into the chat. It answers your customers in seconds from your own docs, says so when it doesn’t know, and shows you what they ask about most, and its answers spend none of your Lovable credits. The free plan gives you 50 answers a month, needs no card and has no end date.

What building a support chatbot in Lovable takes

Lovable’s guide to building an AI chatbot says a first version for one focused job can often be built in 30 minutes or less, and that training it to answer well is an ongoing process. The chat window is the quick part. A chatbot that handles support needs four more pieces, and Lovable writes each one when you ask.

1. Your help content in the app’s database, turned into embeddings so the chatbot can find the right passage for each question. Lovable’s page on AI features calls this semantic search and retrieval-augmented generation (RAG), and lists the embedding models it offers for it.
2. Instructions that keep the answers to that content, and a reply for when it has nothing. Lovable’s guide tells you to set clear fallback behaviour, handing off to a human or giving contact details.
3. A way for you to hear about it when someone needs a person, such as an email, and somewhere to reply.
4. A record of what people asked. The same guide says to review unanswered questions weekly, because they show the gaps in what the chatbot knows.

Each piece is another prompt, paid in build credits, and yours to look after when something changes. Lovable’s AI page does give you a view of the calls, under Cloud › AI, with each request’s model, tokens, cost and status. It doesn’t group the questions or count who asked them.

What a Lovable chatbot costs to run

Lovable’s credits guide says one credit balance pays for building your app, chatting with Lovable, hosting the app with its built-in backend, and the AI features your published app runs. Each model call your chatbot makes takes credits by the model used and the tokens it handles, so longer chat histories and more retrieved passages cost more.

The grant set aside for those calls is 4 credits a month on the Free, Pro and Business plans, and Lovable calls it a temporary offering. After that, your chatbot draws on the same general credits you build with. Pro starts at $25 a month for 100 credits, or $21 a month billed yearly, and a Pro top-up costs $15 for 50 credits.

When the balance reaches zero, the chatbot stops. The credits guide says AI features in published apps stop working while your balance is at zero, and apps that use the built-in backend pause shortly after, with their database, storage and sign-in. Free workspaces can’t buy top-ups, so their chatbot waits for the grant to come back on the 1st of the month. A paid workspace can buy a top-up or turn on auto top-up.

You can’t point the built-in AI connector at your own OpenAI key either. Lovable’s AI page says the connector always bills your workspace credits. To use your own key, you ask Lovable to call the provider from a backend function instead, with the key stored as a secret.

What Deacon does on a Lovable app

You built your app with Lovable so you could ship fast. Answering the same questions by hand slows you down again, and a chatbot you build yourself competes with your app for the same credits. Deacon answers your users in seconds, at any hour, from the docs and pages you choose and in whatever language each person writes in. Its answers come from your Deacon plan, so a busy week of questions never leaves you short of credits to build with.

When your docs don’t cover a question, Deacon says it doesn’t know rather than making something up, and asks for the user’s email. If they leave it, or ask for a person, you get an email. Reply from Deacon’s dashboard, and if they still have the chat open, your reply appears there under your name.

Then Deacon gets better. Every question your docs didn’t cover waits in the dashboard with an Answer this button, and what you write becomes part of what Deacon knows. It can answer the next person who asks, so the more you answer, the less often you’re needed.

You also hear from everyone who asks, not only the few who email. Every conversation is saved and searchable by what users asked. Topics groups the product questions, counts how many people asked each one and labels most of them, such as a bug report, a how-to or a feature ask. That’s the weekly review Lovable’s guide recommends, done for you, and it shows what to fix and what to build next.

Every feature is on every plan, the free one included, and there’s no badge to pay to remove. Nothing is billed past your plan’s answers, no plan upgrades itself, and every plan answers with the same model, so no credits run out faster on a better one. The pricing page has the paid plans.

Adding Deacon to a Lovable app

It takes a few minutes, and Deacon and Lovable has every step.

1. Sign up, then add your docs or your site in Knowledge Sources.
2. On the Install page, list the address your app is published on. A free Lovable app lives on lovable.app, since custom domains need a paid Lovable plan.
3. Copy your script tag from the Install page, paste it into the Lovable project chat and ask Lovable to add it just before the closing body tag on every page. Lovable’s guide to embeds says to paste embed code into the chat and ask Lovable to place it.
4. Publish, then open the published app. Deacon’s Install page shows when it has seen your script on a live address, with a Check again button.

Questions about Lovable chatbots

Can Lovable build a customer support chatbot?

Yes. Describe it in the project chat and Lovable builds the chat window, the backend function and the database tables, with its AI connector running the model. Keeping the answers to your docs, saying so when it doesn’t know and telling you when someone needs a person are pieces you ask for and then look after yourself. Replit chatbots and Bubble chatbots cover building one on those platforms.

Does a Lovable chatbot work on the free plan?

Until its credits run out. Free workspaces get a 4-credit monthly grant for AI features in published apps and can’t buy top-ups, so once the grant and any bonus credits are spent, the chatbot’s calls fail until the grant comes back on the 1st of the month. Lovable’s Pro plan starts at $25 a month for 100 credits.

Can I use my own OpenAI key in a Lovable chatbot?

Not with the built-in AI connector, which always bills your Lovable credits. Ask Lovable to call the provider from a backend function with your key stored as a secret, and the provider bills you instead. The function itself still runs on your Lovable credits.

Can I keep a chatbot I built in Lovable and add Deacon?

Yes. Deacon is a script tag and changes nothing in your app’s code or database, so what you built keeps working. Keep your own chatbot for things only your app can do, such as working with a signed-in user’s records, and let Deacon answer everyone’s questions from your docs. Deacon never looks into a user’s own account or data, and declines those questions.

Does Deacon use my Lovable credits?

Only for the one message that asks Lovable to add the tag, which is build usage like any other. After that, Deacon’s answers run on Deacon’s servers and come from your Deacon plan, so your Lovable credits go on your app.

What happens when Deacon’s free answers run out?

The chat keeps working. A user who asks gets a short reply saying Deacon can’t look into that one right now, and is asked for an email unless they’ve already left one in that conversation. You can still read and reply to every conversation, nothing is billed and your plan doesn’t upgrade itself.