Repository navigation
Merge latest bkt3 (bkmain b7af1e60) into stage - #327
Merged
Merged
Conversation
…gister Google's MCP connector registers more than 5 redirect URIs, and the register endpoint refused it with invalid_redirect_uri. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fix(server): accept up to 20 MCP OAuth redirect URIs so Google can register
…10-11 feat(mcp): agent toolkit, scheduled prompts and session watches (promote to bkmain)
…ns block A Codex async question is a runtime request answered by message. The shell leaves those out of pendingRuntimeRequest and only flags them, so the agent toolkit's loader read nothing: t3_inbox showed the question without options and t3_answer said nothing waits. The loader now reads pending message-mode requests whenever the shell flags one. Every question shows whether it blocks the agent; t3_answer takes a short requestId, reports how a non-blocking answer reached the agent, and dismisses a non-blocking question with dismiss: true. t3_send_message no longer refuses for a non-blocking question. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…rs only its question Review fixes: - an approval next to a non-blocking question loaded no details, because the approval read sat in an else branch; - t3_answer without requestId now refuses when more than one question is open, and every answer hint carries the short requestId, so a retried answer cannot land on the next question; - a secret request (the shell stands it in as a question) now says that a person enters it in the app, instead of asking to try again; - t3_inbox counts the non-blocking questions it does not show; - the delivery check reads only the answer's own run. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…the secret id type - the secret stand-in id is compared as a string (TS2367: TurnItemId vs RuntimeRequestId); - t3_answer refuses a question, dismiss or decision that has a clientRequestId but no requestId, so a retry cannot land on a request asked since; approval hints carry the short requestId; - t3_send_message names a secret request instead of pointing to t3_answer; - secret text in JSON output is cut like other agent-written text. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fix(mcp): t3_answer sees Codex async questions and says which questions block
chore: promote stage to bkmain (Running shelf, collapse-all, Claude subagent end fix)
…rver) TEC-1306 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
chore: promote stage to bkmain (TEC-1306 settled shelf follows the server)
- Session time-in-state tracker: one state per session (agent_running, waiting_human, waiting_workers, queued, idle, archived) emitted as session.state.span events, plus run/request/subagent/plan events and a 5-minute concurrency snapshot. Same reducer drives a history backfill. - Identity: distinct_id is the environment user's work email, instance property and group, property allowlist that drops free text. - No upstream default PostHog key; telemetry stays off without one. - Agent child processes no longer inherit the analytics env. - Server events for MCP tool calls, scheduled tasks, webhooks, session recovery and Linear links. - Web/desktop client analytics via lazily loaded posthog-js, named events only, config delivered by the server at runtime. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- No zero-length spans; span id includes state, waitKind and run id. - Ignore late updates of finished runs; close only active spans at shutdown. - Backfill: normalised emails, --until required for --send. - Web: no persistent opt-out, strip all URL-derived properties, identify only by normalised email. - Register backfill script with knip; scope one Effect diagnostic. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
feat(server): bkt3 product metrics in PostHog
A thread kept its PLAN pill, Plan Ready group and needs-human count for as long as its plan stayed active, even after plan mode was turned off. A plan now waits only while its thread is in plan mode, matching upstream's Plan Ready rule, so stale tags clear without a migration. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit e925a5b)
- Remove the plan card's Preview button; Expand plan stays. - New header: icon, title, state, step count, reading time, and a Widen plan toggle that lets cards use the free width beside the chat. - Plan bodies get the chat's markdown features (skills, runnable shell commands, image preview, artifact templates). - HTML in a plan renders as a live themed page, like html_render: a whole HTML plan, a raw HTML document, or an html-render fence. - Selecting plan text offers Cite and the comment actions; highlights, pins and scroll-to-quote work on plans, and the panel labels them. - Plan-mode turns tell the agent how to add an html-render visual. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit 0d14438)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit a0d2255)
- The plan-format hint joins the fork's plan policy only when plan submission is off, so provider replays keep their recorded input. - Compare a citation's message id with the plan id as plain strings. - Type the comments panel's message list so plans can join it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit 9e3b3f8)
- Plan HTML loads inline through srcDoc, so a frame opened on its own cannot run with the app's origin. The reader's theme is written into the page for the first paint and posted on change. - A remounted plan page starts at its last measured height, so the virtualised list does not jump. - A wide card starts after the timeline minimap's clickable strip. - Plan cards read a memoised slice of the timeline state. - Comment gutter bars sit in the card padding, not over the text. - Settings and docs no longer point at the removed Preview button. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit 0901743)
…smissed When a run fails while the agent waits on a question (the provider stream closes, an event fails to ingest), the run ends and its provider session is detached, but the question stays pending with a live capability. Submit then failed with "Provider session ... was not found", dismiss failed with "This question needs an answer", and Stop was gone. A request whose session is detached or whose run has ended is now handled in message mode: an answer resolves it and reaches the agent as the next message, a dismissal closes it, and an approval closes without a reply. The composer marks such a question dismissible. Multi-select answers join their choices when sent as a message. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit e8ee65f)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit 2a6abc8)
- A completed run no longer counts as ended: its session stays live and a background subagent can still wait on an approval. Failed, interrupted and cancelled runs do. - Read only the request's own run, and take it from the timeline item when the node has none. - MCP lists a question whose run failed or stopped as non-blocking, so t3_answer can dismiss it and t3_send_message no longer refuses. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit 56e1306)
…ime out) (#318) * fix(server): fast webhook ack and personal-token toolyard proxy Session webhook receive: the admission transaction called the Toolyard integration's owner check, which reads environment users through Effect.runPromise on a fresh fiber. The SQLite client has one connection behind a one-permit semaphore held by the open transaction, so that read waited for the transaction that waited for it. Every push callback hung until Toolyard's 20 s client timeout, the request was interrupted and the insert rolled back. The transaction now runs SQL only; the binding is checked before it and again by the dispatch loop. Upstream MCP proxy: a personal T3 token (t3usr_) may call /mcp/upstream/toolyard as the token's own user, with that user's managed Toolyard credential, when the server and user external MCP switches are on. No x-t3-session-id is sent for these calls. Provider sessions are unchanged. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * test(server): fix the session-webhook owner-check test for lint and the live clock - Reuse one payload for the duplicate delivery. Under it.live a second h.event() carries a new timestamp, so the retry was a 409 content conflict instead of a duplicate. - Keep the fresh-fiber Effect.runPromise read, which replicates the production owner check, behind a reasoned oxlint suppression. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * revert(server): drop the personal-token toolyard proxy from this PR The host relay timer that needed a personal t3usr_ token on /mcp/upstream/toolyard was cancelled (2026-10-15), so the new access path has no user. Keep only the fast session-webhook ack. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
…ezing bkt3 The minute pull-request sweep checked all 1,100 unsettled threads on bkt3. Each check starts 6-7 git processes and every start forks the 1.4 GB server on the event loop, so a sweep ran 50-95 s of every two minutes with the main thread saturated (55% of samples in uv_spawn). Replies took 504 ms median during a sweep against 2 ms outside it, and desktop connections hit their 15 s setup limit. A thread with no run or message for a day is now checked once every 30 passes; its own events still queue an immediate check. The agent toolkit's plan-flag correction read a run's turn items with an `IN (...) OR (...)` filter, so SQLite walked every item of the thread in ordinal order: 0.3-2.6 s per cold 30,000-item thread, measured as a 2.5 s event-loop stall at 10:32 UTC. Run filters now read through the (thread_id, run_id) index: 4 ms median on the same cold threads. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit 5afbacb)
fix(server): promote the pull-request sweep and plan-flag freeze fix to bkt3
Tushar, 2026-10-19: pull the latest bkt3 before the stage -> bkmain merge, and deploy to stage first. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
github-actions
Bot
added
vouch:trusted
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
This PR merges the latest
bkmain(b7af1e60, production BkT3) intostage(f073c8c0), on Tushar's instruction of 2026-10-19: pull the latest bkt3 before the stage → bkmain merge, and deploy to stage first. After this merge, stage runs production's code plus the stage-only work. The stage → bkmain promotion (#324) then carries only that stage-only work.The merge has no conflicts. It adds 46 files (+5545 / −78) from
bkmaintostage:bkmaindeploy/bkt3/posthog-dashboards.py)stage)stage)Stage-only work that stays: #317 (only the newest plan stays actionable), #319, #320, #322 (Claude account revive queue and rebalance), #325.
Migrations: the merged registry ends at 1053 (
ThreadClaudeAccountWait), with no duplicate ids.How I tested
git merge-treereported no conflicts, and the merge commit's tree matches it.~/.t3/stage-dev/deployed-shaequals the merge commit and that stagebkt3 serves it.Risks and rollback
stage.🤖 Generated with Claude Code (Claude Opus 5.5, T3 Code)