Universal Discovery Interface Chatbot Prototype

This repository contains the prototype for the Universal Discovery Interface (UDI) chatbot.

The chatbot responds to queries with visualization of the available datasets. This interface is useful as we develop and experiment with the UDI-Grammar.

🚅 Quick start for developers

Install the dependencies

yarn
# or
npm install

Start the app in development mode (hot-code reloading, error reporting, etc.)

quasar dev

Lint the files

yarn lint
# or
npm run lint

Format the files

yarn format
# or
npm run format

Build the app for production

quasar build

Customize the configuration

See Configuring quasar.config.js.

set environment variables (optional)

Variable	Description	Default
VITE_LLM_API_BASE_URL	Base URL for the LLM API.	http://localhost
VITE_LLM_API_PORT	Port where the LLM API is running.	55001
VITE_DATA_PACKAGE_PATH	Path to the data package that lists all data resources available to visualize.	./data/hubmap_2025-05-05/datapackage_udi.json
VITE_PRODUCTION	Toggles production mode, set to true to hide debug features.	false
VITE_BENCHMARK_ENDPOINT_URL	Path or endpoint for benchmark analysis results.	./data/benchmark/benchmark_analysis.json
VITE_AUTH_TOKEN	Token for validation on the backend.	""
VITE_ENABLE_CUSTOM_API_KEY	When true, allows users to provide their own API key via the UI.	false
VITE_CUSTOM_API_KEY_VALIDATION_URL	Endpoint used to validate user-provided API keys.	https://api.openai.com/v1/models

Example .env file:

# .env
# LLM API configuration
VITE_LLM_API_BASE_URL=http://localhost
VITE_LLM_API_PORT=8080

# Data configuration
VITE_DATA_PACKAGE_PATH=./data/my_awesome_data/datapackage.json

# Mode toggle
VITE_PRODUCTION=true

# Enable user-provided API key input in the chat UI
# VITE_ENABLE_CUSTOM_API_KEY=true