The flat single-spawn task wire schema carries arktype `"+": "delete"`, so a
batch `{ context, tasks[] }` payload sent while `task.batch` is disabled has
those keys stripped and is then rejected as `task must be a string (was
missing)` in the agent loop. That preempts the tool's own actionable checks
(validateShapeParams / validateSpawnParams), so the model only ever saw the
misleading arktype error instead of "task.batch is disabled…".
Mark TaskTool with lenientArgValidation so the agent loop forwards the raw
args to execute() on any arktype failure, letting the tool's shape checks
surface the real reason. Valid calls still normalize through arktype; the
success path is unchanged. Mirrors the existing yield-tool pattern.
Fixes#6039
Applied the console-aware launch policy in the original OMP process before spawning the broker. The broker now inherits the caller's console state instead of probing a detached console it created itself.
Applied console-aware spawn options to both non-PTY daemon paths. Windows children now stay attached when the host has a console and use CREATE_NO_WINDOW only for headless hosts.
Fixes#6018
The exact-bounds rewrite matched reverted PR #5812; the ±context expansion
(1 leading + 3 trailing line) around explicit selectors is intended behavior
so edit anchors at range boundaries stay fresh.
- tools.test.ts still asserted the pre-#5812 ±context expansion for offset/
limit and archive-entry reads; updated to the exact-bounds contract.
- selector-controller-logout.test.ts mocked the old modelRegistry.refresh;
#5786 switched logout to a provider-scoped refreshProvider(id, 'online'),
so the mock never resolved and the test timed out.
- Shift+Enter on a tree entry forks with a branch summary directly, skipping
the summary prompt and ignoring the branchSummary.enabled gate.
- Plain Enter keeps existing behavior: direct switch, with the summary prompt
only for users who enabled branchSummary.enabled.
- Updated the selector help line and added controller-level regression tests.
Fixes#5152
Rebased the committed-row seam when multiplexer reflow exposes native scrollback inside an enlarged viewport.
Added production transcript coverage for a finalized prefix above a live response and verified no ED3 is emitted.
Fixes#6011
Native combinators (AbortSignal.any, fetch) brand-check internal slots
that a Proxy cannot forward; return signals unwrapped from
bindBrowserRunFacade so tab.signal composes with native cancellation.
Main renamed/refactored #tryShakeRescueForDeadEnd + #emitShakeRescueNotice into
the tiered #rescueCompactionDeadEnd (elide, then image drop, notices emitted
internally), so a textually-clean merge of this branch left calls to undefined
private methods. Re-express the !preparation rescue through the new API:
progress = prepareCompaction succeeding on the rewritten branch, skipElide when
falling through from a shake pass (the image tier still gets a chance), and
historyRewritten flagged whenever a tier freed content even without progress.
Restore the baseline dead-end remedy text (image drop is automated now, so the
manual /shake images suggestion is stale) and align the rescue-refit regression
test with main's continuation gating (auto-continue after compaction now
requires an active goal or queued work).
The v10 model cache never persists request headers (#5780). The registry's
startup cache readers (#loadCachedStandardProviderModels /
#loadCachedDiscoverableModels) read cache.models directly, so cached rows
replaced bundled models (github-copilot, kimi-code, nanogpt) with header-less
copies for the cache TTL after every restart. Restore bundled static headers,
drop unrestorable rows so bundled fallbacks win the merge, and mark
discoverable providers stale when header-bearing models were excluded.
refreshProvider's #reloadStaticModels could also evict models discovered by
other runtime providers; re-merge them from cache with online-if-uncached.
SettingsSelector dispatches schema settings by full path (terminal.showImages),
so the showImages case never fired from the /settings flow. Add the full-path
case label and cover both ids in the side-effect test.
roboomp