* refactor(agents): split claude-live-session into concept modules
* refactor(agents): delete duplicated live-session helpers
* fix(agents): fence live-session close during pending spawn
* chore(lint): ratchet max-lines baseline after live-session split
* style(agents): satisfy lint on live-session split modules
* fix(agents): fence live session close state
* refactor(agents): extract Claude live turn timeouts
* fix(agents): preserve exec policy after Claude live split
* test(agents): complete Claude live policy session fixture
* refactor(agents): move announce family into subagents/announce/
* refactor(agents): move completion and main-session recovery families
* refactor(agents): move registry family into subagents/registry
* refactor(agents): update registry state type imports
* style(agents): format moved subagent imports
* test(agents): preserve isolated registry test routing
* fix(scripts): recognize relocated subagent announce seams
collection-rollback's directory helper lost its external caller in
bdf202ccc8 — make it local; register collection-review's focused-test
seam per the knip ignoreIssues convention; regenerate the API baseline
after recent harness/channel contract drift.
* refactor(channels): flatten channel turn dispatch naming
* docs(plugin-sdk): narrow inbound reply compat guidance
* docs(channels): point stale references at turn defining modules
* fix(channels): preserve dispatch contracts after flattening
* chore(plugin-sdk): ratchet surface budgets after flattening
* chore(channels): ratchet removed export collisions
* fix(plugin-sdk): restore inbound reply compat exports
Restore eight still-existing legacy callable re-exports from canonical SDK seams and cover the deprecated package subpath with a table-driven compatibility test.
Raise the public export, callable export, and deprecated export budgets by exactly eight; the three maintainer-authorized zero-consumer symbols remain removed.
* test(channels): split channel turn kernel coverage
Replace the oversized kernel test with independently mocked delivery, pipeline, and finalize suites, preserving all 51 tests while removing the max-lines suppression and stale ratchet entry.
* chore(plugin-sdk): refresh inbound reply API hash
* fix(ci): align channel turn review fixes
Restore the test-local DeliveryResult type removed during the split.
Ratchet the public export, callable export, and deprecated export budgets by exactly seven: six channel-inbound plus one channel-outbound legacy re-export.
* feat(ui): float task-suggestion cards top-right with copy-prompt action
- Float a compact tray in .chat-main__conversation with absolute positioning, min(400px, 100% - margins) sizing, elevation, and internal scrolling.
- Remove the icon column, keep dismiss in the header row, and move a compact split button to the bottom-right.
- Add Copy prompt to the menu via shared copyToClipboard, copying the raw prompt while display-only sanitization remains documented inline.
Release note: Suggested-task cards now float as compact cards at the chat top-right and offer Copy prompt in the action menu.
* style(ui): neutral grey task-suggestion card and actions
Drop the accent-tinted card border/background, mute the eyebrow, and use default grey button treatment for the split control per operator direction.
* fix(ui): ungate copy-prompt and surface clipboard outcomes
- Menu now always renders with the client-local Copy prompt while acceptance-mode items stay capability+scope gated (addresses ClawSweeper finding 1).
- Copy success shows a transient Copied label and failures surface through the pane error path (finding 2).
- Task-suggestion pane logic extracted to chat-pane-task-suggestions.ts and the tray adapter moved into the component to keep files under the 700-line cap.
* fix(ui): test task suggestions through tray adapter
* test(ui): cover task-suggestion clipboard failure
* test(ui): split task-suggestion pane coverage
* chore(ui): raise startup JS budget for task-suggestion card feature
QA-profile measurement was 324617-324620 B versus the 324608 B effective budget; this owner-directed PR #121259 bump follows the baseline file’s documented feature-growth precedent.
* chore(ui): raise startup JS ceiling to 318 KiB for task-suggestion cards
The fixed 317 KiB ceiling (324608 B) was within ~20 B of CI-measured startup bytes after accumulated main drift; the task-suggestion card feature crossed it. Ceiling raised with in-file maintainer-approval comment per its own convention; ratchet baseline set to CI-measured 324626 B.
* fix(ui): keep task-suggestion copy available read-only
Remove the stale accept-or-dismiss tray guard now that Copy prompt is a client-local action. Keep Gateway-backed accept and dismiss controls independently gated, with a regression for operators who can only list suggestions.
* test(ui): cover list-only task-suggestion copy
Update the mocked-Gateway E2E to match the repaired read-only behavior: the suggestion remains visible, its mutating Start action is disabled, and Copy prompt stays available.
* fix(ui): scope the Control UI cursor convention to app-like display modes
The app-chrome cursor convention from #103357/#103411 was applied
unconditionally, so an ordinary browser tab lost the pointer hand on
buttons, menus, tabs, rails, selects and accordion summaries — the only
hover affordance a page owns. The convention is correct for the installed
window the manifest declares ("display": "standalone"), not for every
window the same bundle is served into.
base.css now owns one policy token, --cursor-action, selected by display
mode: pointer by default, the desktop arrow under standalone, minimal-ui
and window-controls-overlay. A low-specificity rule maps generic
actionable controls onto that token, restoring the affordance on the
surfaces #103411 stripped bare, while every component rule that owns a
semantic cursor (not-allowed, disabled, grab, resize, zoom-in, text,
wait) keeps winning without !important.
The 92 cursor: pointer declarations that had drifted back into ui/src
since July now consume the token instead of hardcoding the hand, so they
stop contradicting the policy in an installed window. Real hyperlinks
keep the pointer in every mode. The pre-boot mount fallback repeats the
policy locally because it must render when the bundle fails to load.
Closes#121242
* test(ui): run the cursor policy browser test in the node-driven project
ui/vitest.config.ts routes Playwright-from-Node .browser.test.ts files to the
unit-node project; without registering the new cursor policy test there, the
in-browser chromium project tried to import it and failed on node:fs/playwright.
* fix(ui): keep the desktop arrow in the native app hosts
The macOS dashboard embeds the Control UI in a plain web view, which
reports `display-mode: browser`, so the display-mode-only policy would
have handed it the browser-tab pointer. It already announces itself with
`openclaw-native-macos`/`-nav`/`-web-chrome` on `<html>`, the same markers
`ui/src/styles/layout.css` matches on, so the policy reads those too.
* fix(subagents): wake the parent when a follow-up finishes a yielded child
A sub-agent that calls sessions_yield on its own behalf parks its run and
correctly withholds the parent's announce. But a later follow-up to that same
child session registered a sibling registry row instead of continuing the paused
one, so the requester defaulted to the child's own main session and the original
parent — itself idle behind sessions_yield — was never woken. The paused row also
stayed an unsettled descendant, deferring the parent's settle batch forever with
nothing recorded explaining the silence.
Follow-up dispatch now adopts the paused row through the existing post-steer
replacement seam, inheriting the requester identity and carrying the settle-wake
credential forward with its frozen batch membership remapped to the new run id.
A follow-up that names its own requester keeps registering separately, since an
explicit requester is a delivery opt-in that adoption would silently drop.
Also stops frozen-result refill from targeting paused rows: a yield clears the
result on purpose, so refilling from the session would attribute a later turn's
text to the paused run.
Closes#120157
* fix(subagents): select the paused owner past a requester-bound sibling
Adoption looked up the newest run for the child session and adopted it only
when that row was itself paused. A requester-bound follow-up deliberately stays
a sibling, but it registers at a higher generation and becomes that newest row,
so any later default follow-up saw an unpaused newest row, declined adoption,
and registered yet another sibling. The original requester stayed parked behind
a paused row that can never announce -- the same silent stall this fix exists to
remove, reached through a valid mixed-delivery sequence.
The latest-run query now takes an optional predicate applied before the
generation comparison, so a caller that owns a specific row class selects the
newest row of that class. Adoption asks for the newest `sessions_yield` row
directly instead of inferring it from generation order.
Docs now state that continuation applies to default delivery, since a follow-up
carrying its own requester runs as a sibling by design.
* test(qa): prove post-yield follow-up delivery through the gateway boundary
The unit and gateway-method tests for paused-run adoption assert on registry
rows, which proves the bookkeeping but not that an operator ever sees the
result. This adds the boundary proof: a real gateway child, the QA mock channel,
and the mock provider driving a subagent that pauses itself and finishes only on
a later follow-up.
A fixture plugin owns both legs. Its `before_dispatch` hook spawns the child with
`completionDelivery: "current-requester"`, so the announce has the operator turn
as its audience. An HTTP route then dispatches the follow-up to that same paused
session using default delivery -- the path adoption is meant to catch. A
requester-bound follow-up would opt into its own audience and run as a sibling
instead, so the two legs must differ here.
The mock provider gains a child that yields on its own behalf. Both of its turns
match on the current prompt rather than the shared transcript, so the yielded
kickoff cannot make the follow-up turn yield a second time.
The scenario asserts both sides of the invariant: no outbound traffic while the
child is paused, and exactly one announce carrying the follow-up marker once it
ends.
Reverting the adoption call site fails this test in the way that matters: the
child still produces its marker and the run still ends with stopReason=stop, but
nothing reaches the requester and the wait times out. The result is computed and
then silently dropped -- which is the failure this repair exists to remove.
* fix(ci): match QA Lab fixture plugin entries as a group in knip
The all-exports pass listed one fixture entry by name, so every new QA Lab
fixture plugin lands as an unused file and turns check-dependencies red until
someone remembers this file. Nothing imports these entries by design: the
Gateway E2E loads them through plugin config paths.
* docs(subagents): scope yield continuation to plugin runtime follow-ups
Adoption is gated on plugin_subagent task tracking, which only
createGatewaySubagentRuntime().run sets, so api.runtime.subagent.run is the
sole route into it. Writing that as one example implied other follow-up paths
to a paused session continue the run too; they are not tracked as sub-agent
runs and announce nobody.
* fix(subagents): reject undurable paused-run adoption
Fail plugin follow-up admission closed when the paused-run ownership swap cannot be persisted, while retaining the existing restart-recovery return-false contract. Trim duplicate tests and keep boundary coverage for requester routing, wake-batch remapping, repeated yield, and persistence rollback.
Co-authored-by: zhou.huanfeng <woundfongv3@163.com>
* docs(subagents): clarify yielded-run steering
Co-authored-by: zhou.huanfeng <woundfongv3@163.com>
---------
Co-authored-by: Peter Steinberger <steipete@gmail.com>
* refactor(ui): remove fake token estimate
* feat(ui): split model picker into model and effort popovers
* fix(ui): hide filtered-out model rows under the flex option style
* fix(ui): drop unused model locator in cloud-startup e2e
* fix(ui): restore focus and expose active model option to assistive tech
Peter Steinberger