[kixelated]

Problem

release has no committed JavaScript API reports, so trunk's release API diff has no baseline for the first cut. Implements the release backport requested in moq-dev/moq#6477, tracked by moq-dev/moq#6852.

Approach

Backport js/common/api.ts unchanged from trunk and call it from the shared published-package build. Generate every report from release's own sources and exports. just js fix rebuilds with MOQ_API_REPORT=update; normal builds refuse missing, stale, or orphaned reports.

Impact

• Public package APIs and wire formats: unchanged. No source declarations or package versions change.
• Builds check committed js/*/api/*.api.md reports; just fix regenerates them.
• Add the same API Extractor 7.59.4 dev dependency used on trunk.

Alternatives

Copying trunk's reports would baseline a different API. Backporting trunk's whole API-diff change would bring unrelated source and CI changes into release; the cut already carries the release diff workflow.

Validation

• MOQ_STRICT=1 just ci check origin/release: passes in the supplied Nix dev-shell environment.
• MOQ_STRICT=1 just ci test origin/release: passes.
• All 30 typed entrypoints across 14 published release packages have reports.
• Stale and missing report refusal, orphan cleanup, regeneration, and unchanged report acceptance: pass.
• Trunk's API-diff script with API_DIFF_BASE=HEAD --deny js: all 30 release reports unchanged.
• Generator is byte-for-byte identical to trunk; no package source declarations changed.

Follow-ups

Merge the quest completion PR moq-dev/moq#6852 after this backport lands. No merge is authorized by this task.

(Written by GPT-6)

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-26T06:00:11.657205Z	e44d51f	Draft marked ready

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[kixelated]

Automated review by review (OpenAI)

Reviewed commit: e44d51f

No actionable bugs found in the inspected build integration and report generator. The narrow backport is a sensible direction: seed reports from release’s own API rather than copying trunk’s baseline, reuse the existing generator, and enforce freshness in the shared published-package build. The 30 report paths match the typed exports of the 14 affected packages; normal checking and explicit regeneration also account for orphaned reports.

Verification limits: static GitHub inspection of the changes, package manifests, surrounding packaging/declaration checks, and sample reports only. I did not run builds/tests or independently regenerate all reports; the validation results in the PR description remain author-reported.

[kixelated]

Merge summary for e44d51fecd675c96935d6aae8f215c177344987f:

• Changes: backports js/common/api.ts unchanged from main, checks API reports in the shared published-package build, and commits 30 API Extractor reports generated from release's own sources. No source declarations, package versions, or wire formats change.
• Review: Codex completed its review of this head with no findings. The AI review posted from the maintainer account also found nothing actionable. No unresolved threads.
• CI: all 18 required checks pass. quest check --base origin/release passes.
• Decisions: none outstanding. This lands as a merge commit, per CONTRIBUTING's rule for PRs into release.

(Written by Claude Opus 5.5)