fix(agents): handle overloaded failover separately (#38301)

* fix(agents): skip auth-profile failure on overload

* fix(agents): note overload auth-profile fallback fix

* fix(agents): classify overloaded failures separately

* fix(agents): back off before overload failover

* fix(agents): tighten overload probe and backoff state

* fix(agents): persist overloaded cooldown across runs

* fix(agents): tighten overloaded status handling

* test(agents): add overload regression coverage

* fix(agents): restore runner imports after rebase

* test(agents): add overload fallback integration coverage

* fix(agents): harden overloaded failover abort handling

* test(agents): tighten overload classifier coverage

* test(agents): cover all-overloaded fallback exhaustion

* fix(cron): retry overloaded fallback summaries

* fix(cron): treat HTTP 529 as overloaded retry

src/commands/agent.ts

@@ -174,7 +174,7 @@ function runAgentAttempt(params: {
   primaryProvider: string;
   sessionStore?: Record<string, SessionEntry>;
   storePath?: string;
-  allowRateLimitCooldownProbe?: boolean;
+  allowTransientCooldownProbe?: boolean;
 }) {
   const effectivePrompt = resolveFallbackRetryPrompt({
     body: params.body,
@@ -325,7 +325,7 @@ function runAgentAttempt(params: {
     inputProvenance: params.opts.inputProvenance,
     streamParams: params.opts.streamParams,
     agentDir: params.agentDir,
-    allowRateLimitCooldownProbe: params.allowRateLimitCooldownProbe,
+    allowTransientCooldownProbe: params.allowTransientCooldownProbe,
     onAgentEvent: params.onAgentEvent,
     bootstrapPromptWarningSignaturesSeen,
     bootstrapPromptWarningSignature,
@@ -868,7 +868,7 @@ async function agentCommandInternal(
             primaryProvider: provider,
             sessionStore,
             storePath,
-            allowRateLimitCooldownProbe: runOptions?.allowRateLimitCooldownProbe,
+            allowTransientCooldownProbe: runOptions?.allowTransientCooldownProbe,
             onAgentEvent: (evt) => {
               // Track lifecycle end for fallback emission below.
               if (

src/commands/models/list.probe.test.ts

@@ -9,6 +9,7 @@ describe("mapFailoverReasonToProbeStatus", () => {
   it("keeps existing failover reason mappings", () => {
     expect(mapFailoverReasonToProbeStatus("auth")).toBe("auth");
     expect(mapFailoverReasonToProbeStatus("rate_limit")).toBe("rate_limit");
+    expect(mapFailoverReasonToProbeStatus("overloaded")).toBe("rate_limit");
     expect(mapFailoverReasonToProbeStatus("billing")).toBe("billing");
     expect(mapFailoverReasonToProbeStatus("timeout")).toBe("timeout");
     expect(mapFailoverReasonToProbeStatus("format")).toBe("format");

src/commands/models/list.probe.ts

@@ -106,7 +106,7 @@ export function mapFailoverReasonToProbeStatus(reason?: string | null): AuthProb
     // surface in the auth bucket instead of showing as unknown.
     return "auth";
   }
-  if (reason === "rate_limit") {
+  if (reason === "rate_limit" || reason === "overloaded") {
     return "rate_limit";
   }
   if (reason === "billing") {