fix: classify mixed billing fallback summaries

altaywtf · altaywtf · commit 2ea757ce8c8d · 2026-05-20T20:35:13.000+03:00
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -72,6 +72,7 @@ Docs: https://docs.openclaw.ai
 - Control UI/usage: truncate long context skill, tool, and file names in the usage panel while keeping the full name available on hover. (#42197) Thanks @Rain120.
 - Codex: respect explicit `models auth order set` and `config.auth.order` precedence over stale `lastGood` in `/codex account`, and show `no working credential` when every explicit-order profile is ineligible instead of marking a lower-ranked profile as active. Fixes #84386. (#84412) Thanks @openperf.
 - Agents: honor `messages.suppressToolErrors` for mutating tool failures so configured chat surfaces do not receive separate warning payloads. (#81561) Thanks @moeedahmed.
+- Agents/fallback: surface billing guidance for mixed rate-limit plus billing fallback exhaustion instead of generic failure copy. Fixes #79396. (#79489) Thanks @aayushprsingh.
 
 ## 2026.5.19
 
diff --git a/src/auto-reply/reply/agent-runner-execution.test.ts b/src/auto-reply/reply/agent-runner-execution.test.ts
@@ -3549,7 +3549,7 @@ describe("runAgentTurnWithFallback", () => {
     });
   });
 
-  it("does not show a rate-limit countdown for mixed-cause fallback exhaustion", async () => {
+  it("surfaces billing guidance for mixed-cause fallback exhaustion", async () => {
     state.runWithModelFallbackMock.mockRejectedValueOnce(
       Object.assign(
         new Error(
@@ -3593,7 +3593,7 @@ describe("runAgentTurnWithFallback", () => {
 
     expect(result.kind).toBe("final");
     if (result.kind === "final") {
-      expect(result.payload.text).toBe(GENERIC_RUN_FAILURE_TEXT);
+      expect(result.payload.text).toBe("billing");
       expect(result.payload.text).not.toContain("All models failed");
       expect(result.payload.text).not.toContain("402 (billing)");
       expect(result.payload.text).not.toContain("Rate-limited");
diff --git a/src/auto-reply/reply/agent-runner-execution.ts b/src/auto-reply/reply/agent-runner-execution.ts
@@ -380,7 +380,7 @@ function buildRateLimitCooldownMessage(err: unknown): string {
   if (codexUsageLimitMessage) {
     return codexUsageLimitMessage;
   }
-  if (isFallbackSummaryError(err) && isPureBillingSummary(err)) {
+  if (isFallbackSummaryError(err) && hasBillingAttemptSummary(err)) {
     return BILLING_ERROR_USER_MESSAGE;
   }
   const message = formatErrorMessage(err);
@@ -455,11 +455,11 @@ function isPureTransientRateLimitSummary(err: unknown): boolean {
   );
 }
 
-function isPureBillingSummary(err: unknown): boolean {
+function hasBillingAttemptSummary(err: unknown): boolean {
   return (
     isFallbackSummaryError(err) &&
     err.attempts.length > 0 &&
-    err.attempts.every((attempt) => attempt.reason === "billing")
+    err.attempts.some((attempt) => attempt.reason === "billing")
   );
 }
 
@@ -632,7 +632,7 @@ export function buildKnownAgentRunFailureReplyPayload(params: {
   const message = formatErrorMessage(params.err);
   const isFallbackSummary = isFallbackSummaryError(params.err);
   const isBilling = isFallbackSummary
-    ? isPureBillingSummary(params.err)
+    ? hasBillingAttemptSummary(params.err)
     : isBillingErrorMessage(message);
   if (isBilling) {
     return markAgentRunFailureReplyPayload({
@@ -2270,7 +2270,7 @@ export async function runAgentTurnWithFallback(params: {
       }
       const message = formatErrorMessage(err);
       const isBilling = isFallbackSummaryError(err)
-        ? isPureBillingSummary(err)
+        ? hasBillingAttemptSummary(err)
         : isBillingErrorMessage(message);
       const isContextOverflow = !isBilling && isLikelyContextOverflowError(message);
       const isCompactionFailure = !isBilling && isCompactionFailureError(message);
