OpenAI O1 Code Interpreter

Simple hack to add code interpreting capabilities to OpenAI o1. It's able to display charts as outputs as well.

All code execution runs on e2b; you can modify #code_interpret to run locally instead (not recommended for safety reasons).

Setup

• Add API keys to .env based on .env.sample
• Modify the code with a new prompt or pass it through the cli: python main.py Create a report on the evolution of labor productivity in the United States, and what major technological shifts led to it
• Outputs are saved to o1_outputs folder

TODOs:

• When a chart is created, automatically save it and match the name in the report. Right now I save the charts manually in the folder.