Repository navigation
Promote full stage to bkt3 (2026-10-21, includes #337, #338–#340) - #341
Merged
Merged
Conversation
…#15629) Co-authored-by: Julius Marminge <julius0216@outlook.com>
…gdotgg#14581) Co-authored-by: Ennio Cuteri <enniocuteri@fluence-group.com> Co-authored-by: Julius Marminge <julius0216@outlook.com>
…dotgg#18723) Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: Julius Marminge <julius0216@outlook.com>
…gg#17118) Co-authored-by: Julius Marminge <julius0216@outlook.com>
…ce (pingdotgg#19715) Co-authored-by: Julius Marminge <julius0216@outlook.com>
…ine, except recording stop (pingdotgg#16941) Co-authored-by: hogeheer499-commits <267467744+hogeheer499-commits@users.noreply.github.com> Co-authored-by: Julius Marminge <julius0216@outlook.com>
…otgg#14753) Co-authored-by: Julius Marminge <julius0216@outlook.com>
…otgg#12768) Co-authored-by: Rohan Patnaik <rohan-patnaik@users.noreply.github.com> Co-authored-by: Julius Marminge <julius0216@outlook.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
…t the project (pingdotgg#18692) Co-authored-by: Claude <noreply@anthropic.com> Co-authored-by: Julius Marminge <julius0216@outlook.com>
…o the bottom (pingdotgg#17065) Co-authored-by: Julius Marminge <julius0216@outlook.com>
… second (pingdotgg#21242) Co-authored-by: Julius Marminge <julius0216@outlook.com>
…nd viewport (pingdotgg#19913) Co-authored-by: Julius Marminge <julius@t3.gg> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> Co-authored-by: Julius Marminge <julius0216@outlook.com>
…ab (pingdotgg#13608) Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Co-authored-by: Julius Marminge <julius0216@outlook.com>
… child thread (pingdotgg#18362) Co-authored-by: Julius Marminge <julius0216@outlook.com>
…pingdotgg#19636) Co-authored-by: Julius Marminge <julius0216@outlook.com>
…ed history (pingdotgg#21188) Co-authored-by: Julius Marminge <julius0216@outlook.com>
…tgg#16688) Co-authored-by: Julius Marminge <julius0216@outlook.com>
…nute cap (pingdotgg#19649) Co-authored-by: Julius Marminge <julius0216@outlook.com>
…pingdotgg#20054) Co-authored-by: Andrey Velts <a.velts@vk.team> Co-authored-by: Julius Marminge <julius0216@outlook.com>
…me folder (pingdotgg#20950) Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: Julius Marminge <julius0216@outlook.com>
…t page (pingdotgg#19724) Co-authored-by: FranzCh <28222983+Fr4nzz@users.noreply.github.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: Julius Marminge <julius0216@outlook.com>
…tion (pingdotgg#21053) Co-authored-by: Julius Marminge <julius0216@outlook.com>
…er (pingdotgg#19072) Co-authored-by: Ryan Wible <197168665+rywible@users.noreply.github.com> Co-authored-by: Julius Marminge <julius0216@outlook.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
…g#19947) Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: Julius Marminge <julius0216@outlook.com>
…ingdotgg#15836) Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: SAPHID <saphid@users.noreply.github.com> Co-authored-by: Julius Marminge <julius0216@outlook.com>
…ed (pingdotgg#17458) Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: Julius Marminge <julius0216@outlook.com>
…tured (pingdotgg#20521) Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> Co-authored-by: Julius Marminge <julius0216@outlook.com>
…18827) Co-authored-by: SAPHID <saphid@users.noreply.github.com> Co-authored-by: GPT-6.1 Sol <noreply@openai.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: Julius Marminge <julius0216@outlook.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
…cache (pingdotgg#22712) Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
…g#22717) Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
…rok commands stop holding runs open (pingdotgg#22709) Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: Fuksss <fuxik12345@gmail.com>
68 upstream commits on top of dac3437, including upstream's test pruning and test-hygiene bans. Upstream migration 067 (WebhookDedupe) registers at fork id 1059. The plan-review Mermaid block now uses upstream's pele-based MermaidDiagram (upstream dropped mermaid). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Re-export resolveChangeRequestPresentation and deriveLogicalProjectKey, which upstream stopped exporting; the fork phase sidebar and smart-git button use them. - formatWorkingDurationLabel moved to client-runtime models; the phase sidebar's snoozeWakeDescription lives in a fork module now that upstream dropped it. - Upstream's new orphaned follow-up MCP test gets the fork scope fields; the archive undo test expects the fork's duplicate-archive guard. - Formatting for apps/web/package.json and PlanReviewNodes. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Fork tests follow upstream's new test-hygiene rules (no real-clock or frame waits, no wall-clock reads). - The skipped upstream 063 test is removed; the fork keeps 063 unregistered. - The mobile plan-comment test re-imports countReviewCommentContexts. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Merge upstream main (163bdea) into stage
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
… shell On bkt3 (2026-10-21) the server logged 98 event-loop stalls in 24 h once the Claude accounts hit their limits. ForkProviderSessions.listSessions read the full shell (active and archived) on the event loop and then one record read per thread, about 2.2 s, and maybeRebalance called it at least twice at every turn end of a thread on a full account before checking for room. - listSessions reads only the threads live sessions serve, from a new in-memory ProviderSessionManager.listLive (fork-marked, optional). - maybeRebalance checks for room elsewhere before any session read. - A released Claude session also settles its process-bound subagents when a replacement session already serves the thread (the hard-limit path places the replacement before the old release finishes; 9 cards stayed running). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Review follow-up: a session manager without listLive (a wrapper or a Layer.mock) silently brought back the blocking full-shell path. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…shows Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…ell-reads fix(server): account policies no longer freeze the server reading the shell
…ds one thread - ProjectionStore.getShellSnapshot runs on the SQLite reader worker, so the remaining writer-path callers (worktree branch rename, MCP agent session lists, manager.sessions, startup) no longer block the event loop for ~2 s. - threadUsage.get reads only the requested thread's shell row instead of the full active shell on every open thread view each minute. - AGENTS.md: rules for server code that must scale with bkt3's database. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…ute and listLive fallback Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The resting composer row overlays the effort, attach and queue icons on the full placeholder text. Show "Ask anything..." while resting and keep the full hint in the expanded composer. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…uards fix(server): full shell reads leave the main thread, thread usage reads one thread
…placeholder fix(web): compact composer shows a short placeholder
github-actions
Bot
added
vouch:trusted
This was referenced Oct 29, 2026
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.
Promotes the full
stage(e65c637) tobkmain, so bkt3 production runs it.This branch is built on top of #337 (
promote/stage-2026-10-21, head 629f121) and adds the stage merges that #337 lacked:Merging this PR also merges #337, because #337's head is an ancestor.
How I tested
Not tested here. Tushar asked to merge the full stage at once with no extra tests (2026-10-21). The tree is identical to
origin/stage, which stagebkt3 already serves. #337's CI was green on its head.Requested by: tushar
🤖 Generated with Claude Code