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/agents/auth-profiles/types.ts

@@ -39,6 +39,7 @@ export type AuthProfileFailureReason =
   | "auth"
   | "auth_permanent"
   | "format"
+  | "overloaded"
   | "rate_limit"
   | "billing"
   | "timeout"

src/agents/auth-profiles/usage.test.ts

@@ -177,6 +177,24 @@ describe("resolveProfilesUnavailableReason", () => {
     ).toBe("auth");
   });
 
+  it("returns overloaded for active overloaded cooldown windows", () => {
+    const now = Date.now();
+    const store = makeStore({
+      "anthropic:default": {
+        cooldownUntil: now + 60_000,
+        failureCounts: { overloaded: 2, rate_limit: 1 },
+      },
+    });
+
+    expect(
+      resolveProfilesUnavailableReason({
+        store,
+        profileIds: ["anthropic:default"],
+        now,
+      }),
+    ).toBe("overloaded");
+  });
+
   it("falls back to rate_limit when active cooldown has no reason history", () => {
     const now = Date.now();
     const store = makeStore({

src/agents/auth-profiles/usage.ts

@@ -9,6 +9,7 @@ const FAILURE_REASON_PRIORITY: AuthProfileFailureReason[] = [
   "billing",
   "format",
   "model_not_found",
+  "overloaded",
   "timeout",
   "rate_limit",
   "unknown",
@@ -35,7 +36,7 @@ export function resolveProfileUnusableUntil(
 }
 
 /**
- * Check if a profile is currently in cooldown (due to rate limiting or errors).
+ * Check if a profile is currently in cooldown (due to rate limits, overload, or other transient failures).
  */
 export function isProfileInCooldown(
   store: AuthProfileStore,
@@ -508,7 +509,7 @@ export async function markAuthProfileFailure(params: {
 }
 
 /**
- * Mark a profile as failed/rate-limited. Applies exponential backoff cooldown.
+ * Mark a profile as transiently failed. Applies exponential backoff cooldown.
  * Cooldown times: 1min, 5min, 25min, max 1 hour.
  * Uses store lock to avoid overwriting concurrent usage updates.
  */