context-labs / autodoc

Experimental toolkit for auto-generating codebase documentation using LLMs

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Allow for configuration of indexing strategy.

samheutmaker opened this issue · comments

Autodoc currently only supports indexing a file using the most affordable models available in a projects autodoc.config.json. Ideally, we should allow for different types of indexing strategies. At very least, there should be a option to use the "best available" models, which would tell Autodoc to always choose the most powerful models that have been configured.

This is an evolving issue. Please reach out on Discord if you're interested in contributing.

Related #8.

commented

It's too bad I don't know how to integrate TS with Python, otherwise I could make some suggestions using gpt-index/llama-index. There's another project called Cognosis AI that is the only other GPT tooling project I've seen.