alyt

I might be misunderstanding what you're asking, but VSCode already makes regions foldable by default.

That said, my extension does support regions in more languages than VSCode currently supports, so maybe you're thinking of a particular language in which VSCode doesn't make regions foldable, that my extension could fill the gap for?

Btw, another extension I'd personally recommend is Region Highlighter by 'Wiensss', which makes regions easier to see in the editor itself by coloring them, and also provides a command for making regions (although it is limited in language support). I don't currently use any other region extensions.

Region Highlighter: https://marketplace.visualstudio.com/items?itemName=Wiensss....

I think it's 0 right now lol

> Every time I need to specify output schema for LLMs, I need to write a converter from Pydantic models to the schema. Pydantic V2's model_json_schema is not supported by some platforms like OpenAI or n8n.

Interesting idea; I'm not familiar with n8n, but doesn't OpenAI's API natively support Pydantic models? You can just pass in the model directly as the `response_format` argument: https://platform.openai.com/docs/guides/structured-outputs?a...

Definitely possible I'm missing which use cases you're thinking of. Would be cool to see some more concrete examples of how this has been useful for you!

Would be really cool if it could just keep generating new regexes to solve for! So it could be like a training game.

And also a version that involves making a regex that passes two sets of test strings (one set for strings that should match the regex, one for strings that shouldn't match).

Nice design btw, minimalistic but satisfying. The two features above would make this a really sick project overall IMO

> Does this actually work? Yeah, somewhat! Could it create scripts that erase your drive? Maybe! Good luck!

Love the transparency lol.

Cool idea though! I imagine the "erase your hard drive" risk could be mitigated pretty well by adding an extra LLM step to double-check the generated script and explain each part of it (maybe add an `# explanatory comment` at the end of each line), and then an extra human step to confirm before executing the script.