JPT - ChatGPT Code Interpreter Chrome 插件, crx 扩展下载

                    
ChatGPT's code interpreter plugin as a chrome extension
JPT let's you run your Python code right within ChatGPT. No need for ChatGPT Plus or any ChatGPT plugins!

JPT uses Pyodide, a WASM-based Python interpreter. This means that your code executes completely within your browser. The only people who get to see your code is you and OpenAI!

Features
- Run Python code
- Edit Python code, but does not persist
- Upload files for Python to access, including Python modules
- Download files created in Python
- Standard library, numpy, matplotlib, and some other packages supported
- See output from stdout
- See plots from matplotlib

How to use
1. To run Python code, ask ChatGPT to write a hello world program in Python. Then press the "Run" button. It may take a while for the first run, but subsequent runs should be faster.
2. To edit Python code, press the "Edit" button. Once you are done editing, press "Edit" again to save.
3. To upload a file, press the "Upload" button and select the file you want to upload. You can then access the file in the current directory in Python.
4. To download a file, first save the file within Python and remember the name you saved the file as. Press the "Download" button, you will be prompted to save a file. Type in the name of the file you saved earlier, including the file extension. It must match exactly or it will not download. I'm aware this is not very user-friendly and I will make a proper UI in future versions.
5. To load your own module, upload your Python file. Then load the module dynamically. You can ask ChatGPT to give you an example of how to do this.
6. To load Python packages, use micropip. For example, to import scipy, use `micropip.install("scipy")`. You may also need to import micropip. More details here: https://pyodide.org/en/stable/usage/loading-packages.html

Known Issues
- Infinite loops will freeze JPT. Requires restarting the extension
- Input is not supported

Roadmap
- Interrupt execution
- Add indicator for initialization and execution
- Support input
- Export chat as Jupyter Notebook
- Reduce extension size