Fully switch to oxlint

[david-crespo]

Oxlint runs in 800ms without caching. Eslint runs in 14 seconds without caching, 600ms with a warm cache and no changes. So Oxlint without caching is about the same as eslint with caching.

$ time npm run lint

> oxide@0.0.0 lint
> eslint --ext .js,.ts,.tsx app test mock-api

npm run lint  21.42s user 2.09s system 166% cpu 14.150 total

$ time npm run lint -- --cache

> oxide@0.0.0 lint
> eslint --ext .js,.ts,.tsx app test mock-api --cache

npm run lint -- --cache  0.69s user 0.22s system 83% cpu 1.083 total

$ time npm run lint -- --cache

> oxide@0.0.0 lint
> eslint --ext .js,.ts,.tsx app test mock-api --cache

npm run lint -- --cache  0.67s user 0.17s system 144% cpu 0.579 total

We've been running both eslint and oxlint for a while. Now that oxlint supports type-aware rules and JS plugins, I think it's time to drop eslint. The playwright plugin runs via oxlint's experimental JS plugin support. We do lose eslint-plugin-react-hook-form for now because all 3 rules use context.getScope(), which oxlint doesn't support yet. I'll make an issue to add it back when it's supported.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Review	Updated (UTC)
console	Ready	Preview	Dec 24, 2025 0:01am