Integrate error message back to LLM

continuedev / continue

⏩ Continue is the leading open-source AI code assistant. You can connect any models and any context to build custom autocomplete and chat experiences inside VS Code and JetBrains

https://docs.continue.dev/

Apache License 2.0

14.8k stars 1.08k forks source link

Integrate error message back to LLM #1268

Open crudalex opened 3 months ago

crudalex commented 3 months ago

Validations

[X] I believe this is a way to improve. I'll try to join the Continue Discord for questions
[X] I'm not able to find an open issue that requests the same enhancement

Problem

The initial code generated by LLM is not always perfect and requires manually feeding the error message back to LLM for refinement.

Solution

The plugin should be able to capture error messages and feed the error messages back to the LLM to generate a fix.

sestinj commented 3 months ago

@crudalex I very much agree with this principle. When you say "error message" in this case, are you thinking of error messages in the terminal, or lint errors (red/yellow underlines in VS Code editor)?

Mostly curious what workflow you're finding yourself in most often

crudalex commented 3 months ago

Actually both. But the issues in the problem panel are the first ones that I will want to resolve. Next step probably integrate into stack trace errors?

sestinj commented 3 months ago

great, yeah that makes sense. today you can use cmd/ctrl+shift+R to automatically debug errors in the terminal, but taking this further is a good idea