This application aims to provide a template for building generative UI applications with LangChain.js. It comes pre-built with a few UI features which you can use to play about with gen ui. The UI components are built using Shadcn.

First, clone the repository and install dependencies:

git clone https://github.com/bracesprou/gen-ui.git

cd gen-ui

yarn install

Next, if you plan on using the existing pre-built UI components, you'll need to set a few enviroment variables:

Copy the .env.example file to .env:

The OPENAI_API_KEY is required. LangSmith keys are optional, but highly recommended if you plan on developing this application further.

Get your OpenAI API key from the OpenAI dashboard.

Sign up/in to LangSmith and get your API key.

Create a new GitHub PAT (Personal Access Token) with the repo scope.

Create a free Geocode account.

# ------------------LangSmith tracing------------------
LANGCHAIN_API_KEY=...
LANGCHAIN_CALLBACKS_BACKGROUND=true
LANGCHAIN_TRACING_V2=true
# -----------------------------------------------------

GITHUB_TOKEN=...
OPENAI_API_KEY=...
GEOCODE_API_KEY=...

To run the application in development mode run:

yarn dev

This will start the application on http://localhost:3000.

To run in production mode:

yarn start

yarn build

If you're interested in ways to take this demo application further, I'd consider the following:

• A custom LangGraph agent instead of the default AgentExecutor amd createToolCallingAgent
• Adding a "classifier" step before the tool call, with a small fast model (e.g Claude 3 Haiku) which selects the tool call/component first. This would improve overall latency for the time to first UI (with the loading state component).