Adaptive Model Routing

Use when selecting or switching models dynamically during a session to conserve tokens, choosing model tiers for subtasks, configuring lifecycle hooks for model gating, or dispatching subagents with matched reasoning density.