mirror of
https://github.com/stablyai/orca.git
synced 2026-11-15 18:37:24 +00:00
Brennan Benson
abc8850180
* Unify dedicated agent launch choices and argument badges
* Enable action model and effort menus with argument migration
* Bound argument choice edit range processing
* Preserve independent action effort through launch and migration
* Preserve literal config arguments and explicit launch permissions
* Respect repository action argument groups and share launch controls
* Remove duplicate action launch command coverage
* Add agent setup search and controls to the new tab menu
* Show inherited choices beside picker reset options
* Keep terminal Arguments separate from shared launch choices
* Launch default agents from rows and select resolved setup values
* Keep terminal settings available while Chat defaults load
* Reuse existing launch setup controls and Arguments editor
* Unify Source Control action choices and repository overrides
* Use precise types in action choice contracts
* Consolidate action choice test imports
* Correct action choice types and integration test boundaries
* Complete prepared terminal test result contract
* Align action integration expectations with canonical choices
* Keep launch defaults authoritative and couple CLI controls
* Assert production Chat preparation runs in launch contract
* Preserve alternate Chat launches and bound picker argument derivation
* Publish committed picker priority and satisfy control contracts
* test(mobile): type the file preview retry test helpers so the test typecheck passes (#35431)
(cherry picked from commit 0774fc9373)
* fix: keep picker integration tests within compiler boundaries
* fix: revalidate completed picker defaults on reopen
* test: give exported launch fixture mocks portable signatures
* fix: keep fallback launch carriers out of action mode controls
* Preserve Fast argument badges outside dialog mode controls
* Project existing Fast carriers independently of model capabilities
* Keep setup focus during launch row pointer transitions
* Allow declared tier choices without an inherited selection
* fix: place agent setup beside the launch menu
* Commit edited action arguments before planning and launch
Reconcile deliberate raw edits once into the submitted action recipe so replacement model and effort flags reach Save and Start together. Preserve captured sources and untouched load conflicts.
* Name portable Source Control dialog fixture results
* Keep agent setup contained and preserve known argument policy
* Use checked style properties in popup geometry fixture
* Keep setup scrolling inside a plain content wrapper
* fix(agent-picker): match the setup panel to the approved layout
Flush model search and list, one settings section with label-left rows
(Effort, compact Permissions menu, Fast mode), a fixed Arguments
disclosure row with its hint, and a summary footer with Launch chat /
Launch terminal. Settled defaults with nothing chosen keep the model
list and Effort pickable. Hovering another menu row no longer closes
the open panel.
* fix(agent-picker): use primitive-owned spacing and update launch labels in tests
* fix(agent-picker): add panel labels to localization catalogs
* Keep one launch setup open and align picker spacing
* Tighten launch model rows and clarify terminal hint
* Keep keyboard focus in replacement launch setup
* fix(tab-bar): keep a long agent setup result's chevron inside the menu
* test(agent-picker): wait for outgoing focus cleanup event
* Show empty model search feedback and name setup dialog
* test: register selection catalog in launch guard fixture
* fix(agent-picker): retain setup drafts and contain Arguments
* fix(agent-picker): retain authored setup route after alternate launch
* style(agent-picker): use a ghost setup chevron with divider
* style(agent-picker): apply muted color to setup glyph
* test(agent-picker): declare retained catalog default