OpenHands
OpenHands copied to clipboard
Reviewing code changes using a linter.
What problem or use case are you trying to solve? When models are writing code to modify files they often make spacing mistakes and other errors that can be easily detected using a linter.
Describe the UX of the solution you'd like Performance enhancement for all agents!
Do you have thoughts on the technical implementation? Use a docker image within the current environment that will be able to compile and test the code for any simple errors.
Additional context This is in conjunction with the SWE-Agent implementation #846
I have discussed this with @guneetsk99 and he is going to work on this.
Can we do this generically? How do you plan to detect the correct language, and choose the right linter?
stubs here indicate linter selection is by file extension, and a desire to support 4 languages.
FWIW, the original swe-agent implementation only lints python.
This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.