[aldoeliacim]

Summary

• Add regression coverage for unchanged Chat Completions messages and Responses input histories: reuse the frozen transformed history, preserve current request controls, and avoid another compressor call.
• No production code changes. The existing implementation already handles this case; this is coverage, not a demonstrated fix for an upstream safety refusal.

Related Issues

• Retains the scope and review history of test(compression): verify unchanged-history prefix reuse #18304. duplicate: compression history coverage → #18304 #18305 and fix(providers): a provider safety refusal is not an account ban #18307 remain closed duplicates.
• ⚠️ base-red inherited: 🔴 Release branch not green: release/v3.8.53 #17219

Target Branch

• Base is the active release/v3.8.53 branch, reconciled at 139070680f77728e2309eb4df18463282a1c990a.
• This is test coverage, not a new feature.

Validation

• Change type: other — compression regression coverage.
• node --import tsx/esm --test tests/unit/compression/live-zone.test.ts tests/unit/compression/live-zone-current-turn.test.ts — 124 passed, including both new cases; rerun after commit.
• npm run lint — passed on the exact base and reconciled candidate.
• Unchanged pre-commit and commit-message hooks — passed. The UI-kit snapshot check took its configured public-fork missing-clone path; it did not verify the private clone.
• npm run check:changelog-integrity and npm run check:test-tla-order — passed.
• npm run check:build-scope — passed on base and candidate.
• npm run check:docs-counts and npm run check:mutation-test-coverage remain red on the exact base and candidate with byte-identical diagnostics: seven documentation-count drifts and five missing covering-test entries across four mutated modules. No suppressions or baseline changes are included.
• Exact-head hosted build, security tests, Semgrep, Vitest and the no-new-ESLint check passed. Aggregate CI remains red; skipped coverage is not a pass.
• The failed unit jobs identify docs-count synchronization, the SSO direct-main guard, and environment-documentation contracts. Exact-base API/dashboard typecheck and error-helper runs reproduce the same failures. The dependency manifests are unchanged; the hosted vulnerability ratchet remains red. The local triage audit was skipped because its scanner was unavailable, so no local vulnerability result or advisory-feed comparison is claimed.
• Existing candidate base repairs are tracked in fix(release): clear the remaining base-reds on release/v3.8.53 (docs counts, SSO entry guard, API diagnostics, vuln ratchet) #18658, fix(ci): keep timed model-sync logs visible to the error-helper gate #18711, fix(quality): restore provider asset provenance validation #18602 and fix(ci): restore missing mutation test coverage links #18157; model-sync environment documentation is also present in [defer] feat(providers): integrate native Rosebud asset generation #18577. These are overlap/dependency references, not approval of those PRs or a waiver of required checks. No duplicate base-repair implementation or baseline increase is included here.

Tests Added Or Updated

• tests/unit/compression/live-zone.test.ts: two parameterized unchanged-history cases.
• No production code changed.

Coverage Notes

• Covers existing live-zone prefix reuse for both supported sequence fields, changing streaming/output limits while retaining the cached history, and zero live items without recompression.
• Full coverage measurement is left to CI; no coverage percentage is claimed here.

Reviewer Notes

• Retains the removal of the redundant production guard and the withdrawal of the earlier empty-message/upstream-safety causal claim.
• The candidate excludes the unrelated Gemini/API/i18n/suppression changes present in the previous head.
• No migrations, feature flags, dependency changes, or runtime changes.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota.

An organization owner can choose the repository organization as the billing source for reviews requested by members with a paid Copilot license. Learn about Copilot code review billing.

[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	⚠️ Failed 2026-10-28T02:52:35.868325Z We couldn’t complete this review. Comment @codex review to try again.	ecb3d9b	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.