bhaskar511939/react-streaming-chat-boilerplate

This project is a ready-made chat interface for building your own AI chatbot app, similar in feel to ChatGPT or Claude's website. Instead of building the chat window, message bubbles, and streaming text effect from scratch, you can start with this and connect it to your own AI backend. The interface shows AI responses appearing word by word as they are generated, rather than waiting for the full answer before showing anything, which matches how most modern AI chat products behave. It also displays small status indicators showing what the AI is doing behind the scenes, such as running a tool or searching the web, and lets users switch between different AI models like GPT-4o or Claude Sonnet in the middle of a conversation. The design uses a dark color theme and adjusts to fit mobile phone screens. To use it, you need your own backend server that can generate AI responses and send them using a streaming technology called Server-Sent Events, in a specific message format this project expects. The README lists the exact structure your backend needs to follow, including how to signal that a tool is running, that new text has arrived, or that the response is finished. It is not tied to one specific backend framework, so it can work with LangGraph, FastAPI, Express, or other servers, as long as they follow that format. The author also maintains a separate companion project that provides a matching backend built with FastAPI and LangGraph. The frontend is built with React, TypeScript, and Vite, and intentionally avoids extra UI libraries or complex styling tools, using plain CSS so it stays simple to read, fork, and adjust. Getting started only requires cloning the repository, installing dependencies, and running the development server, then pointing it at your backend's address in the settings panel. It is released under the MIT license, free to use and modify.