chore: migrate to oxlint

[logaretm]

Benchmarks

Overal

Metric	Before (ESLint)	After (Oxlint)	Speedup
yarn lint:lerna	~133s	~11s	12x*
CI Time	5m 58s	2m 32s	2.4x*

Note that:

• Lerna adds a considerable overhead that eats a lot of the gains we can potentially get.
• CI time is hogged by building types step

SDK Packages

Package	Files	ESLint	Oxlint	Speedup
core	365	9.6s	53ms	181x
browser	136	6.8s	55ms	124x
node	105	6.1s	64ms	95x
node-core	101	6.2s	56ms	111x
nextjs	181	10.9s	79ms	138x
sveltekit	63	6.4s	71ms	90x
opentelemetry	58	4.3s	52ms	83x
cloudflare	43	3.8s	45ms	84x
remix	38	7.1s	42ms	169x
react	39	6.5s	49ms	133x
feedback	38	3.8s	48ms	79x
replay-internal	152	5.6s	38ms	147x
vue	24	4.0s	48ms	83x
svelte	15	4.0s	52ms	77x
angular	12	3.7s	37ms	100x

Dev Packages

Package	Files	ESLint	Oxlint	Speedup
browser-integration-tests	778	10.8s	209ms	52x
node-integration-tests	605	9.0s	291ms	31x
node-core-integration-tests	268	6.2s	74ms	84x
e2e-tests	10	2.6s	44ms	59x
cloudflare-integration-tests	27	2.5s	35ms	71x
test-utils	5	2.4s	21ms	114x
rollup-utils	13	❌ error	22ms	N/A
bundler-tests	3	❌ error	51ms	N/A

What Changed

• All .eslintrc.* files replaced with .oxlintrc.json configs
• Lint scripts updated from eslint . to oxlint .
• 61 ESLint config files removed

Current State

• 0 errors, 45 warnings (mostly complexity - functions with cyclomatic complexity >20)
• All existing rules migrated or documented as gaps

Gaps (things we lost)

• Import sorting (simple-import-sort) - use Prettier
• Some TypeScript rules (member-ordering, naming-convention)
• 4 custom Sentry rules - can be ported via JS plugins later

Kept

• eslint-config-sdk and eslint-plugin-sdk packages (published for SDK users)
• ESLint in e2e test apps (they simulate real user setups)

Docs

I included more details in docs/oxlint-migration-gaps.md

[github-actions]

Codecov Results 📊

---

Generated by Codecov Action

[github-actions]

node-overhead report 🧳

Note: This is a synthetic benchmark with a minimal express app and does not necessarily reflect the real-world performance impact in an application.

Scenario	Requests/s	% of Baseline	Prev. Requests/s	Change %
GET Baseline	8,732	-	9,086	-4%
GET With Sentry	1,636	19%	1,653	-1%
GET With Sentry (error only)	5,932	68%	6,094	-3%
POST Baseline	1,151	-	1,166	-1%
POST With Sentry	521	45%	539	-3%
POST With Sentry (error only)	1,003	87%	1,027	-2%
MYSQL Baseline	3,138	-	3,282	-4%
MYSQL With Sentry	373	12%	393	-5%
MYSQL With Sentry (error only)	2,589	83%	2,680	-3%

View base workflow run