[botandbang]

What and why

Each locallibrary run asked for 6 macOS jobs (lint, build, 3 test shards, coverage), and the Free plan allows 5 at once. In a 21-hour sample, macOS sat at its cap for 160 minutes, and macOS and Windows jobs on pull requests were 31% of all job-minutes.

• Pull requests and pushes to main run on Linux only. A pull request with the new all-os label runs macOS and Windows too. The label is read from the API, so adding it and re-running CI is enough.
• ci.yml triggers on merge_group. A group runs only locallibrary's macOS and Windows job, since its pull requests already passed on Linux and the push to main runs Linux again. last-passed diffs a group against the commit it goes on top of.
• macOS and Windows now get one job each (other-os): clippy, doctests and cargo nextest run, with no coverage or shards. It has a 90-minute timeout, sized from Windows' times in run 41157262485. Coverage and its thresholds are Linux-only now.
• Moved the Linux build, test, hegel and coverage jobs into locallibrary-ci.yml and removed locallibrary-ci-os.yml, which existed so each OS could fail on its own.
• Added locallibrary-cache.yml, which runs other-os nightly on main so its caches get saved there (shared-key: other-os).
• The ruleset gets a merge queue: build at most 2 groups at once, up to 5 pull requests per group, ALLGREEN, merge commits, a 120-minute check timeout.
• A new policy in ruleset.rego: when the ruleset has a merge queue, every gated workflow triggers on merge_group.
• Added the all-os label, and updated the coverage comment in .cargo/config.toml and the docs.

After merging: an admin applies the ruleset (tools/ruleset.sh apply), then re-runs main's CI. Its ruleset check fails until the queue is on.

How it was tested

• tools/policy.sh: 72/72 Rego tests at full coverage, 48/48 conftest checks. The merge_group rule failed with the trigger removed.
• actionlint and zizmor are clean.
• The changes job's OS logic, run for pull requests (with and without the label), drafts, merge groups, pushes, and pull requests with over 3000 files.
• The merge queue flow, the nightly cache save, and the cold-cache Windows job time only show on GitHub.

[github-actions]

Coverage Report for apps/web

Status	Category	Percentage	Covered / Total
🟢	Lines	95.42% (🎯 93%) 🟰 ±0 pp	7911 / 8290
🟢	Statements	94.08% (🎯 91%) ⬆️ +0.01 pp	12047 / 12804
🟢	Functions	94.21% (🎯 92%) ⬆️ +0.03 pp	3125 / 3317
🟢	Branches	84.71% (🎯 80%) 🟰 ±0 pp	5619 / 6633

File Coverage

No changed files found.

Generated in workflow #7 for commit c0c8cbb by the Vitest Coverage Report Action

Size

Compared with main. KiB are 1,024 bytes.

	This pull request	main	Change
Client JavaScript	680.3 KiB	680.3 KiB	0 B (0%)
Client JavaScript, brotli	225.8 KiB	225.8 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 #7 for commit c0c8cbb 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	-1 B (-0%)
Unpacked (dist/)	842.8 KiB	842.8 KiB	0 B (0%)

The merge-base changed after approval.

[github-actions]

Rust coverage

From run 7. Both are held to the thresholds in .cargo/config.toml.

Tests	Lines	Functions	Regions
All	97.39% (84410/86668)	96.53% (13503/13989)	95.67% (185846/194265)
Hegel property tests alone	95.39% (82672/86668)	94.11% (13165/13989)	93.46% (181563/194265)