Resolve the agent's budget-driven model downshift
Resolve which model the agent’s next turn should use given the organization’s AI budget caps — once the org’s realized daily or monthly AI spend reaches its configured cap and the agent has a config.budget_downshift_model set, every subsequent turn is forced onto that cheaper downshift model until the period resets. The agent runtime reads this ahead of every chat turn (internal service token, dual-auth with the public endpoint), mirroring the per-turn complexity cascade at GET /agents//model-route. Never faults on an unset cap or missing downshift model — it reports reason: no_cap_configured / no_downshift_model_configured and keeps the primary model.
Authorizations
Dashboard JWT token from Clerk
Headers
Sandbox opt-in for Clerk-session-authenticated requests. Set to true to route the call through the test-mode pipeline: no real provider delivery, no credits deducted, response meta.test_mode: true. Ignored for live API keys (dv_live_sk_*) — server-to-server clients must use a test-prefixed key (dv_test_sk_*) to exercise sandbox. Test-prefixed keys unconditionally enable sandbox regardless of this header.
true, false Path Parameters
Identifier of the agent whose downshift to resolve.