[botandbang]

What and why

SvelteKit's client awaits the root error page (nodes/1) in start() before it hydrates, but the server only modulepreloads the route's own nodes (sveltejs/kit#17415). Every page fetched nodes/1 only after start.js had run, a round trip behind the largest chunk, before it could hydrate.

That request is also why Lighthouse's LCP check on /projects/halflight failed on a busy machine (2,404 and 2,408 ms against 2,300). On localhost the first paint lands after the modules run, so the simulated LCP counts every request finished by then. When nodes/1 finished first, which a loaded host makes the usual case, the LCP went from 2,101 to 2,252 ms, and one more simulated round trip put it over the budget.

Added: patches/@sveltejs__kit@3.1.0.patch, which adds the root error page's imports to the modulepreloads (the root-error part of sveltejs/kit#17416), registered in pnpm-workspace.yaml with a note to drop it once a release has the fix.

Changed: tests/build.test.ts counts the root error page in the JavaScript every page loads (now about 179 KiB of the 180 KiB budget), since every page downloads it.

Removed: nothing. The Lighthouse config and its 2,300 ms LCP limit are unchanged.

How it was tested

Full lhci collect on a fresh network, 4 times x 3 runs, with 16 busy processes on the machine: /projects/halflight LCP over the 12 runs went from 2,103 to 2,256 ms (8 of 12 at about 2,252) to 2,100 to 2,118 ms, and / and /tools/locallibrary dropped about 150 ms too. Direct Lighthouse runs with 48 busy processes went from a median of 2,254 ms (max 2,261) to 2,102 ms (max 2,160). The simulated FCP rose about 75 ms because lantern counts the extra modulepreload as blocking; FCP isn't asserted and the performance score stayed at 0.98.

Ran pnpm verify, pnpm test:lighthouse and pnpm test:e2e --project chromium in apps/web, plus tools/policy.sh and ast-grep scan.

Checklist

• Tests cover the change (the build budget test and Lighthouse CI on these pages)
• The READMEs, docs/ and --help still say what the code does

[github-actions]

Coverage Report for apps/web

Status	Category	Percentage	Covered / Total
🟢	Lines	95.43% (🎯 93%) 🟰 ±0 pp	7916 / 8295
🟢	Statements	94.09% (🎯 91%) 🟰 ±0 pp	12053 / 12810
🟢	Functions	94.21% (🎯 92%) 🟰 ±0 pp	3126 / 3318
🟢	Branches	84.71% (🎯 80%) 🟰 ±0 pp	5621 / 6635

File Coverage

No changed files found.

Generated in workflow #34 for commit 1df9ec4 by the Vitest Coverage Report Action

Size

Compared with main. KiB are 1,024 bytes.

	This pull request	main	Change
Client JavaScript	680.7 KiB	680.7 KiB	0 B (0%)
Client JavaScript, brotli	226.8 KiB	226.8 KiB	0 B (0%)
Client CSS	89.6 KiB	89.6 KiB	0 B (0%)
Client CSS, brotli	12.9 KiB	12.9 KiB	0 B (0%)

[github-actions]

Coverage Report for apps/localtoolbox

Status	Category	Percentage	Covered / Total
🟢	Lines	100% (🎯 100%) 🟰 ±0 pp	4303 / 4303
🟢	Statements	99.64% (🎯 95%) 🟰 ±0 pp	5862 / 5883
🟢	Functions	99.65% (🎯 95%) 🟰 ±0 pp	1715 / 1721
🟢	Branches	97.72% (🎯 50%) 🟰 ±0 pp	2791 / 2856

File Coverage

No changed files found.

Generated in workflow #34 for commit 1df9ec4 by the Vitest Coverage Report Action

Size

Compared with main. KiB are 1,024 bytes.

	This pull request	main	Change
Packaged (localtoolbox.zip)	366.1 KiB	366.1 KiB	-3 B (-0%)
Unpacked (dist/)	842.8 KiB	842.8 KiB	0 B (0%)

[botandbang]

Closing and reopening so CI runs on the new single-workflow CI against current main. Branch unchanged.

[botandbang]

Closing and reopening so CI runs against current main (with #353's CodeQL fix). Branch unchanged.