An experiment inspired by reading the paper ReAct: Synergizing Reasoning and Acting in Language Models and A simple Python implementation of the ReAct pattern for LLMs by Simon Willison
# Build the docker image
docker build -t gpt-shell .
# Run the docker image
docker run --env OPENAI_API_KEY=sk-xxxxxxx -it gpt-shell
- This code is just an experiment, it will burn your API credits quite fast and could destroy all your data if not run in an empty container/vm.