Skip to content

fix(site): align guarantee cards - #590

Merged
SUaDtL merged 3 commits into
mainfrom
fix/align-guarantee-cards
Aug 3, 2026
Merged

fix(site): align guarantee cards#590
SUaDtL merged 3 commits into
mainfrom
fix/align-guarantee-cards

Conversation

@SUaDtL

@SUaDtL SUaDtL commented Aug 3, 2026

Copy link
Copy Markdown
Collaborator

Summary

  • Align the three guarantee cards by resetting Starlight's inherited block-start margin.
  • Add a regression guard for the card alignment rule.
  • Stabilize repeated statusline-render tests by pinning their unrelated dirty-state probe.
  • Restore fresh-clone release verification with a reachable, byte-identical historical baseline.
  • Advance the shipped CA payload to 2.11.1 with synchronized public version surfaces.

Why

Starlight's layered sibling flow spacing adds a top margin to later card siblings, which staggered the guarantee row. Repeated statusline tests also sampled a latency-bounded Git probe between whole-render comparisons. Separately, the release trace pinned a feature-branch-only commit that remained available in long-lived local clones but could not exist in fresh CI clones.

Verification

  • npm test in site: 512 tests passed.
  • npm run coverage in site: 70.83% lines and 70.19% branches.
  • npm run typecheck in site.
  • npm run build in site: 134 pages built.
  • npm audit --omit=dev --audit-level=high in site: zero vulnerabilities.
  • Canonical repository script suite and 1,308 hook tests passed locally.
  • Release trace: 29 tests passed.
  • Payload gate: 2.11.0 -> 2.11.1 accepted.
  • Focused landing regression: 22 tests passed.
  • Two Sol reviews: PASS with no findings.

Tradeoff

The statusline tests pin git_dirty instead of extending the production 100 ms timeout. This preserves the UI latency boundary while isolating the render contracts (conflict hierarchy: correctness, level 2, over incidental integration coverage).

Review note

A computed-style browser test across responsive breakpoints remains a non-blocking coverage improvement. The current CSS behavior and source-level regression are verified.

SUaDtL added 2 commits August 2, 2026 22:48
The fallback assertion compares whole renders, so the bounded dirty probe could change an unrelated branch marker between calls. Pin that probe to keep the test focused on palette compatibility.
Starlight's sibling flow spacing adds block-start margin to later cards, creating a staggered row. Reset that margin at the card boundary so grid gap remains the only inter-card spacing.

CHANGELOG: Align guarantee cards across responsive layouts.
@coderabbitai

coderabbitai Bot commented Aug 3, 2026

Copy link
Copy Markdown

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository YAML (base), Organization UI (inherited)

Review profile: CHILL

Plan: Pro Plus

Run ID: cfb6eb3f-a119-4685-9c36-15b79362538f

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

  • 🔍 Trigger review

Comment @coderabbitai help to get the list of available commands.

Fresh CI runners could not resolve a feature-branch-only historical SHA, while repeated render tests sampled a latency-bounded Git probe between comparisons. Pin a reachable byte-identical baseline, isolate the live probe in test fixtures, and advance the shipped CA payload version.

CHANGELOG: Restore cross-platform hook CI and stable statusline regression checks.
@SUaDtL
SUaDtL merged commit d30f2f2 into main Aug 3, 2026
45 checks passed
@SUaDtL
SUaDtL deleted the fix/align-guarantee-cards branch August 3, 2026 10:54
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant