Repository navigation
Conversation
Wire doctor scripts and changed-scope pre-push when gui/ changes; keep CI advisory via react-doctor.yml scanning gui/.
Resolve GUI diagnostics (a11y, dialogs, TanStack Query, etc.) and src performance/security findings so full-repo doctor scans report clean.
Derive poll data from React Query instead of mirroring into useState via effects, and sync editable slices with render-time state adjusts.
Use a controlled clock and a wider stale window so the replacement lock is not immediately re-staled and unlinked when same-bytes rewrites leave mtime unchanged.
Owner
|
Thanks for running React Doctor and collecting these improvements. For future contributions, please split tooling, query migrations, page-level UI changes, update integrity, and runtime adapter changes into separate semantic PRs; combining them makes correctness review and rollback difficult. We are preserving your authorship and rebuilding the retained work as a small stacked-PR series based on
|
5 of 6 tasks
Owner
|
The rebuild of this PR has landed on |
8 tasks
kartikkabadi
added a commit
to kartikkabadi/opencodex
that referenced
this pull request
Aug 14, 2026
Port codex-router PR lidge-jun#140: a request the provider answered cannot have exceeded its context window, so the token estimate must never claim it did. estimateTokens gains an optional contextWindow cap. The request log caps the ESTIMATE field (attempt.inputTokenEstimate and the usage-fallback inputTokens) at the routed model's window while the combined inputTokens field keeps its max(reported, estimate) behavior; provider-reported positive counts are never reduced. Kiro auto and unknown adapters/models stay uncapped - a window is never invented. Window sources: KIRO_MODEL_CONTEXT_WINDOWS (src/providers/kiro-models.ts, mirroring adapters/kiro.ts kiroUpstreamContextWindow) and inferCursorContextWindow (src/adapters/cursor/discovery.ts), picked by route adapter in request-log.ts contextWindowForModel. Tests: extended tests/token-estimate.test.ts, added tests/request-log-estimate-cap.test.ts, updated the one request-log test that asserted an above-window estimate.
lidge-jun
pushed a commit
that referenced
this pull request
Aug 14, 2026
Port codex-router PR #140: a request the provider answered cannot have exceeded its context window, so the token estimate must never claim it did. estimateTokens gains an optional contextWindow cap. The request log caps the ESTIMATE field (attempt.inputTokenEstimate and the usage-fallback inputTokens) at the routed model's window while the combined inputTokens field keeps its max(reported, estimate) behavior; provider-reported positive counts are never reduced. Kiro auto and unknown adapters/models stay uncapped - a window is never invented. Window sources: KIRO_MODEL_CONTEXT_WINDOWS (src/providers/kiro-models.ts, mirroring adapters/kiro.ts kiroUpstreamContextWindow) and inferCursorContextWindow (src/adapters/cursor/discovery.ts), picked by route adapter in request-log.ts contextWindowForModel. Tests: extended tests/token-estimate.test.ts, added tests/request-log-estimate-cap.test.ts, updated the one request-log test that asserted an above-window estimate.
agentHits
pushed a commit
to agentHits/opencodex
that referenced
this pull request
Sep 17, 2026
Port codex-router PR lidge-jun#140: a request the provider answered cannot have exceeded its context window, so the token estimate must never claim it did. estimateTokens gains an optional contextWindow cap. The request log caps the ESTIMATE field (attempt.inputTokenEstimate and the usage-fallback inputTokens) at the routed model's window while the combined inputTokens field keeps its max(reported, estimate) behavior; provider-reported positive counts are never reduced. Kiro auto and unknown adapters/models stay uncapped - a window is never invented. Window sources: KIRO_MODEL_CONTEXT_WINDOWS (src/providers/kiro-models.ts, mirroring adapters/kiro.ts kiroUpstreamContextWindow) and inferCursorContextWindow (src/adapters/cursor/discovery.ts), picked by route adapter in request-log.ts contextWindowForModel. Tests: extended tests/token-estimate.test.ts, added tests/request-log-estimate-cap.test.ts, updated the one request-log test that asserted an above-window estimate.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
doctor/doctor:fullscripts, changed-scope pre-push whengui/changes, and advisory GitHub Action CI.src/(perf loops/clones, Zod 4 APIs, update integrity check, related cleanups).react-doctorscans (No issues found).Test plan
bun run typecheckbun test --isolate ./tests/cd gui && bun run doctor:full→ no issuesbun x react-doctor@latest --verbose --scope full→ no issues