fix(auxiliary): detect quota keywords in _is_payment_error and allow fallback for explicit providers

agent/auxiliary_client.py

@@ -2098,7 +2098,9 @@ def _is_payment_error(exc: Exception) -> bool:
     if status in {402, 429, None}:
         if any(kw in err_lower for kw in ("credits", "insufficient funds",
                                            "can only afford", "billing",
-                                           "payment required")):
+                                           "payment required", "quota",
+                                           "too many tokens", "daily limit",
+                                           "tokens per day")):
             return True
     return False
 
@@ -4522,8 +4524,7 @@ def call_llm(
         # Only try alternative providers when the user didn't explicitly
         # configure this task's provider.  Explicit provider = hard constraint;
         # auto (the default) = best-effort fallback chain.  (#7559)
-        is_auto = resolved_provider in {"auto", "", None}
-        if should_fallback and is_auto:
+        if should_fallback:
             if _is_payment_error(first_err):
                 reason = "payment error"
                 # Resolve the actual provider label (resolved_provider may be
@@ -4851,8 +4852,7 @@ async def async_call_llm(
             or _is_connection_error(first_err)
             or _is_rate_limit_error(first_err)
         )
-        is_auto = resolved_provider in {"auto", "", None}
-        if should_fallback and is_auto:
+        if should_fallback:
             if _is_payment_error(first_err):
                 reason = "payment error"
                 _mark_provider_unhealthy(

tests/agent/test_auxiliary_client.py

@@ -923,6 +923,25 @@ def test_no_status_code_no_message(self):
         exc = Exception("connection reset")
         assert _is_payment_error(exc) is False
 
+    def test_429_with_quota_message(self):
+        exc = Exception("Too many tokens per day: quota exceeded")
+        exc.status_code = 429
+        assert _is_payment_error(exc) is True
+
+    def test_429_with_daily_limit_message(self):
+        exc = Exception("daily limit reached, try again tomorrow")
+        exc.status_code = 429
+        assert _is_payment_error(exc) is True
+
+    def test_429_with_too_many_tokens_message(self):
+        exc = Exception("Too many tokens requested today")
+        exc.status_code = 429
+        assert _is_payment_error(exc) is True
+
+    def test_no_status_quota_exceeded(self):
+        exc = Exception("quota exceeded for this model")
+        assert _is_payment_error(exc) is True
+
 
 class TestIsRateLimitError:
     """_is_rate_limit_error detects 429 rate-limit errors warranting fallback."""
@@ -1111,6 +1130,33 @@ def test_429_rate_limit_triggers_fallback(self, monkeypatch):
         # Fallback client should have been used
         assert fallback_client.chat.completions.create.called
 
+    def test_explicit_provider_gets_fallback_on_payment_error(self, monkeypatch):
+        """Non-auto (explicit) providers should still get fallback on payment/quota errors (#26803)."""
+        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
+
+        primary_client = MagicMock()
+        quota_err = Exception("Too many tokens per day: quota exceeded")
+        quota_err.status_code = 429
+        primary_client.chat.completions.create.side_effect = quota_err
+
+        fallback_client = MagicMock()
+        fallback_client.chat.completions.create.return_value = MagicMock(choices=[
+            MagicMock(message=MagicMock(content="fallback response"))
+        ])
+
+        with patch("agent.auxiliary_client._get_cached_client",
+                    return_value=(primary_client, "google/gemini-3-flash-preview")), \
+             patch("agent.auxiliary_client._resolve_task_provider_model",
+                    return_value=("openrouter", "google/gemini-3-flash-preview", None, None, None)), \
+             patch("agent.auxiliary_client._try_payment_fallback",
+                    return_value=(fallback_client, "fallback-model", "nous")):
+            result = call_llm(
+                task="compression",
+                messages=[{"role": "user", "content": "hello"}],
+            )
+        # Even with explicit provider "openrouter", fallback should trigger
+        assert fallback_client.chat.completions.create.called
+
 # ---------------------------------------------------------------------------
 # Gate: _resolve_api_key_provider must skip anthropic when not configured
 # ---------------------------------------------------------------------------