itspreto / baby-code Goto Github PK

requirements.txt contains PyPDF2==3.9.0
Last PyPDF2 version 3.0.1: https://pypi.org/project/PyPDF2/#history

Incorporating different prompts for personality injections

There is this really cool dataset of curated prompts by the community: awesome-chatgpt-prompts

It would be nice to have some of searchable dropdown that lets you pick from one of these items.

These prompt templates could then work hand-in-hand with the Interactive Mode for inferencing (which is still in the plans) and allow for a more tailored experience when using certain tools.

Currently the inferencing experience is very straight forward and manual.

• You enter a prompt and you either get working code that runs or you get an error.
• If you want to debug this error you have to manually extract the generated script, as well as the stderr from the execution. and re-submit it with a new prompt asking the model to troubleshoot and provide a fully working [updated] version of the script.

It would be ideal to have a built-in mechanism that handles this for you through a button click.

• This would allow for quickly debugging on the fly by automatically inferring what the error is from the stacktrace (if there is one) or requesting a brief explanation of the issue through prompting before attempting to debug.

Scripting vs Interactive Modes

Scripting

• This is essentially the standard and only way to use the application currently and allows for simple a very simple chain:
  PROMPT -> SCRIPT -> EXECUTION -> RESULT

Interactive

• Under this mode the model would have access to various tools (through Langchain) that would allow it to perform various tasks such as opening file(s), summarizing, documenting, creating unit tests for code-- the applications are endless.
  PROMPT -> STEPS -> SCRIPT -> EXECUTION -> STEPS -> SCRIPT -> STEPS -> SCRIPT...

Hey! Thanks for doing this, it is fantastic and I have been having a blast running it on a FreeBSD box.

Silly question and probably an easy one that I am simply fumbling on...how do I hook baby-code.py up to an existing llama.cpp server? Rather than having it start its own.

Thanks for throwing this out there and I think the UI is pretty good, btw :D

Cheers,
-matt

Persisted storage

I didn't have time to implement this so the current version does not support any type of persistence for the chats-- meaning any code you get it to generate you will have to get manually saved for future uses.

Proposal

As there is currently no mechanism for caching/storing any sort of session data I'm proposing a vector db solution for implementing embedding similarity search as well as an actual database for storing chats.