Replaces single-path cf-orch allocation with a three-tier strategy: tier 1 task_allocate() (coordinator-driven), tier 2 direct CFOrchClient.allocate() (TaskNotRegistered fallback), tier 3 local LLMRouter. Module-level imports for CFOrchClient and LLMRouter make all three paths patchable in tests without import caching issues. |
||
|---|---|---|
| .. | ||
| __init__.py | ||
| affiliates.py | ||
| llm_planner.py | ||
| llm_router.py | ||
| llm_timing.py | ||
| planner.py | ||
| prep_scheduler.py | ||
| shopping_list.py | ||