ChatbotBuilder is an open-source, web-based chatbot creation platform built with React.js, GraphQL, and OpenAI/DeepSeek AI models. It allows users to design and deploy AI-powered chatbots tailored to various conversational scenarios. By defining chat scope, guidelines, and intents, AI models can be able to communicate with users intelligently and provide open-ended or customized responses.
- Platform-based: Each user can create their own chatbot after logging in.
- Supports OpenAI and DeepSeek models: Choose between OpenAI or DeepSeek for your chatbot's AI model.
- Customizable Guidelines and Intent Flow: Configure guidelines and intent flow tailored to different conversation scenarios.
- Intent Flow Configuration: Based on user queries, you can define different logic execution flows:
- Functional Handler: Write custom code to handle tasks such as API requests, database queries, etc.
- Non-Functional Handler: Provide non-functional responses, such as returning markdown content, formatted lists, images, etc.
- Model Response Handler: Let the model freely respond while setting guidelines to restrict certain topics or content in the answers.
- Shareable Chatbots: Once created, chatbots can be shared via links and embedded in other applications.
ChatbotBuilder can be used in a wide range of scenarios, such as:
- Customer Support – Assist users with order inquiries, cancellations, and general support.
- Travel Assistant – Help users search for trips, book hotels, and get travel recommendations.
- Immigration Consultation – Provide guidance on visa applications, residency options, and document requirements.
You can even create a companion chatbot that is just for you to chat and counsel you.
Want to give it a try?
- Visit https://www.chatbotbuilder.dev to create and try your own chatbot.
- Or check out the "Farm Assistant Chatbot" at https://www.chatbotbuilder.dev/chat/e0426cc7-46e9-48c1-a1dd-374d95751cc0.
- Created an intent flow for specific intent.
- Provided guidelines to the model, allowing the model to answer freely.
- Let model inject custom content into the answer.
- Created an intent flow for specific intent.
- Provided guidelines to the model, let model return a pre-defined markdown content directly.
- Created an intent flow for specific intent.
- Let the model perform interactive communication to obtain the required parameters.
- Provided custom logic and code to handle it, in this demo the chatbot will send a GET request to API endpoint then return to the user.
Refer to design.md
Refer to app/README.md for frontend setup steps.
Refer to api/README.md for API setup steps.
Once the frontend and API are set up, launch the platform and start creating your AI chatbot.
- Configure chatbot guidelines and intent flow to define how it interacts with users.
- Customize responses to specific customer queries for better user experience.
Refer to video recording https://www.youtube.com/watch?v=VCwTL6jz_XU
Once configured, share the chatbot’s link with users to start engaging conversations.
- Open source
- Low-code & intuitive – Easily create AI chatbots without deep technical knowledge.
- Highly configurable – Define chat behavior, restrict responses, and guide AI interactions.
- Flexible integrations – Works with OpenAI, DeepSeek AI, and can be expanded with custom APIs.
Feel free to ask any questions in our slack.