better-chatbot

Better Chatbot is a feature-rich MCP-enabled AI chat platform that aggregates multiple AI providers and a suite of tools into a unified chat experience. It supports multi-provider LLMs (OpenAI, Anthropic, Google, and more), web search, code execution in JavaScript and Python, data visualization, visual workflows, and custom agents. The MCP integration enables you to call tools directly from chat using @toolname, build visual workflows as callable tools, and even run browser automation tasks via Playwright MCP. You can also experiment with real-time voice interactions and collaborate by sharing MCP configurations with your team. To get started, deploy the app, set up your API keys, and begin invoking tools in chat with simple mentions like @web-search or @python-executor to perform tasks, fetch results, or generate code and visualizations.

Prerequisites:

• Git installed
• Node.js (and npm) installed
• Optional: Docker if you prefer containerized setup

1. Clone the repository git clone https://github.com/cgoinglove/better-chatbot.git cd better-chatbot

2. Install dependencies npm install

3)Configure environment

• Create a .env.local (or use your preferred env method) and add: PORT=3000 OPENAI_API_KEY=your-openai-api-key BETTER_AUTH_SECRET=your-secret GOOGLE_GENERATIVE_AI_API_KEY=optional ANTHROPIC_API_KEY=optional GOOGLE_API_KEY=optional GOOGLE_CSE_ID=optional

4. Start the development server npm run dev

5. (Optional) Build for production npm run build npm run start

6. If you prefer Docker, refer to the Docker deployment guide in the repo guides and adapt the container entrypoint to run server.js with the necessary env vars.

Tips and caveats:

• Ensure you supply at least one LLM provider API key (OpenAI, Claude, Gemini, etc.) to enable core functionality.
• The app ships with MCP tools like web search, JS/Python code execution, data visualization, and image generation; you can expose additional tools via visual workflows.
• If you encounter authentication or provider errors, verify environment variables and provider account status.
• For local-first setups, consider enabling local storage drivers and ensuring network access to required APIs.
• The project may have paused development; monitor the repo for issues and security patches, and engage the community for contributions.
• When sharing MCP configurations, redact sensitive keys and only share necessary environment placeholders.