Snowflake + Wikipedia + LLM Demo

This is a small Streamlit app for the Replit Bounty posted by Altimeter Capital. It connects Wikipedia articles and a Snowflake database to GPT so that the model has additional knowledge when answering questions. The data is connected using GPT Index and a LangChain agent is used to determine which data source is used for a specific query.

After putting your OpenAI API key and Snowflake DB info in .streamlit/secrets.toml, run the app locally with:

streamlit run main.py