This project provides a web-based chat interface for interacting with LangGraph agents, targeting developers and researchers building AI-powered applications. It simplifies agent interaction by offering a user-friendly chat UI that connects to any LangGraph server, streamlining the development and testing process.

How It Works

The application is a Next.js frontend that connects to a LangGraph backend. Users input the deployment URL of their LangGraph server, an assistant/graph ID, and optionally a LangSmith API key. The UI then facilitates message streaming and display, allowing for interactive conversations with the agent. It supports hiding messages via specific tags or prefixes for controlled visibility.

Quick Start & Requirements

• Install dependencies with pnpm install.
• Run the app with pnpm dev.
• Requires Node.js and pnpm.
• Deployed version available at agentchat.vercel.app.
• Video setup guide available.

Highlighted Details

• Connects to any LangGraph server with a messages key.
• Supports message streaming and custom message visibility controls.
• Offers two production deployment strategies: API Passthrough and Custom Authentication.
• Environment variables can bypass the initial setup form.

Maintenance & Community

• Developed by LangChain AI.
• Links to LangGraph documentation for advanced setup.

Licensing & Compatibility

• License details are not explicitly stated in the README.

Limitations & Caveats

• Production deployment requires careful configuration of environment variables for authentication and API proxying.
• Custom authentication for production requires additional setup on the LangGraph deployment side.