Typebot — Visual AI Chatbot Builder You Can Self-Host

What it is

Typebot is an open-source visual chatbot builder that lets you create advanced conversational flows using a drag-and-drop interface. It offers 34+ block types for text input, buttons, conditions, API calls, OpenAI integration, payment collection, and more. You embed the chatbot anywhere on your website and collect results in real-time.

Typebot targets marketers, product teams, and developers who need conversational forms, lead capture bots, or AI-powered customer interactions without writing code. It can be self-hosted for full data ownership or used via the hosted service.

Why it saves time or tokens

Building a chatbot from scratch requires handling conversation state, branching logic, UI rendering, and backend storage. Typebot provides all of this visually. The OpenAI integration lets you add AI responses at specific points in the flow without building a full AI backend. For teams already using AI assistants, Typebot provides a user-facing interface that the AI can power behind the scenes.

How to use

1. Self-host Typebot with Docker or use the hosted service at typebot.io
2. Create a new bot and design the conversation flow using the visual editor
3. Embed the bot on your website using the provided script tag or React component

Example

<!-- Embed Typebot on any website -->
<script type='module'>
  import Typebot from 'https://cdn.jsdelivr.net/npm/@typebot.io/js@0.3/dist/web.js'

  Typebot.initBubble({
    typebot: 'my-chatbot-id',
    theme: {
      button: { backgroundColor: '#0042DA' },
      chatWindow: { backgroundColor: '#fff' }
    }
  })
</script>

Related on TokRepo

• AI tools for no-code — no-code AI tools and builders on TokRepo
• AI tools for automation — workflow automation and chatbot tools

Common pitfalls

• Self-hosting requires PostgreSQL and proper environment configuration; the Docker compose file handles this but read the docs before deploying
• OpenAI blocks consume API credits per conversation turn; set token limits and monitor usage to control costs
• Complex branching logic with many conditions becomes hard to maintain visually; break large bots into sub-bots for clarity