feat(studio): message queue and mid-run steering - #980
Closed
jtenniswood wants to merge 1 commit into
Closed
Conversation
The queue machinery first: messages typed while a run drives park as
queuedMessages, drain one per run with flush-on-idle, and render as the
queued strip in the chat view. Steer lands over it: PendingSteer with the
pure splitPendingSteersOnWatermark, orphan requeue, steerMessage/steerQueued/
cancelPendingSteers, and the steer-echo assistant re-anchor. The composer's
resolveComposerAction routes Enter to send/queue/steer, and Personalize
gains the Message-queuing preference (Shift+Enter does the opposite).
Steer is gated on features.has('http_steer') — the daemon endpoints from
earlier in this series — with the explicit queue fallback (too_late, 409
stale_run_control, or an absent capability all queue; text is never
dropped). Against the e2e fixture (no steer capability) the gate stays on
the queue path, so specs are unchanged; the watermark-split and
client-action suites own the steer contract.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
jtenniswood
added
enhancement
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Stack 15/21 — Studio: message queue and mid-run steering
The most smeared feature of the original branch, landed as one coherent delta in dependency order: the queue machinery first, steer over it.
queuedMessages/takeQueued, drain-one-per-run, flush-on-idle;chat-viewgains the QueuedMessageStrip.PendingSteer+ the pure, testedsplitPendingSteersOnWatermark, orphan requeue,steerMessage/steerQueued/cancelPendingSteers, and the steer-echo assistant re-anchor. Gated onfeatures.has('http_steer')— the PR 2 daemon endpoints — with the explicit queue fallback (too_late/409 stale_run_control/ absent capability → queue; the text is never dropped).resolveComposerActiondecides Enter = send/queue/steer; the "Message queuing" preference lands here in Personalize (useEnterSendBehavior; Shift+Enter does the opposite).Steer is deliberately vitest-owned (watermark-split + client-action suites) — the fixture daemon has no steer capability, so e2e exercises the queue fallback path unchanged.
Demo at this stack point: type while a run is driving — messages queue or steer per your preference; cancel a pending steer before it drains.
Part of the stacked landing of #618.
🤖 Generated with Claude Code