From 16a7465b9b513567569ef4cf1a6afbfaf7e21609 Mon Sep 17 00:00:00 2001
From: sgaofen <135070653+sgaofen@users.noreply.github.com>
Date: Wed, 22 Apr 2026 21:19:20 -0700
Subject: [PATCH] fix(agent): inspect wrapped API error bodies

---
 agent/error_classifier.py            | 32 +++++++++++++----------
 tests/agent/test_error_classifier.py | 38 ++++++++++++++++++++++++++++
 2 files changed, 57 insertions(+), 13 deletions(-)

diff --git a/agent/error_classifier.py b/agent/error_classifier.py
index 04875b6a5458..63cdd2e9a82b 100644
--- a/agent/error_classifier.py
+++ b/agent/error_classifier.py
@@ -845,19 +845,25 @@ def _extract_status_code(error: Exception) -> Optional[int]:
 
 
 def _extract_error_body(error: Exception) -> dict:
-    """Extract the structured error body from an SDK exception."""
-    body = getattr(error, "body", None)
-    if isinstance(body, dict):
-        return body
-    # Some errors have .response.json()
-    response = getattr(error, "response", None)
-    if response is not None:
-        try:
-            json_body = response.json()
-            if isinstance(json_body, dict):
-                return json_body
-        except Exception:
-            pass
+    """Walk the error chain to find a structured SDK error body."""
+    current = error
+    for _ in range(5):  # Max depth to prevent infinite loops
+        body = getattr(current, "body", None)
+        if isinstance(body, dict):
+            return body
+        # Some errors have .response.json()
+        response = getattr(current, "response", None)
+        if response is not None:
+            try:
+                json_body = response.json()
+                if isinstance(json_body, dict):
+                    return json_body
+            except Exception:
+                pass
+        cause = getattr(current, "__cause__", None) or getattr(current, "__context__", None)
+        if cause is None or cause is current:
+            break
+        current = cause
     return {}
 
 
diff --git a/tests/agent/test_error_classifier.py b/tests/agent/test_error_classifier.py
index c8faffb0c6d5..28c350940530 100644
--- a/tests/agent/test_error_classifier.py
+++ b/tests/agent/test_error_classifier.py
@@ -120,6 +120,13 @@ def test_from_body_attr(self):
         e = MockAPIError("fail", body={"error": {"message": "bad"}})
         assert _extract_error_body(e) == {"error": {"message": "bad"}}
 
+    def test_from_cause_chain(self):
+        inner = MockAPIError("inner", body={"error": {"message": "nested"}})
+        outer = Exception("outer")
+        outer.__cause__ = inner
+
+        assert _extract_error_body(outer) == {"error": {"message": "nested"}}
+
     def test_empty_when_no_body(self):
         assert _extract_error_body(Exception("generic")) == {}
 
@@ -748,6 +755,37 @@ def test_body_message_enrichment(self):
         # "try again" is only in body, not in str(e)
         assert result.reason == FailoverReason.rate_limit
 
+    def test_wrapped_402_uses_nested_body_for_transient_limit(self):
+        """Wrapped SDK errors should keep the nested body used for 402 disambiguation."""
+        inner = MockAPIError(
+            "Usage limit",
+            status_code=402,
+            body={"error": {"message": "Usage limit reached, try again in 5 minutes"}},
+        )
+        outer = Exception("Usage limit")
+        outer.__cause__ = inner
+
+        result = classify_api_error(outer)
+
+        assert result.status_code == 402
+        assert result.reason == FailoverReason.rate_limit
+        assert result.retryable is True
+
+    def test_wrapped_402_billing_without_transient_signal(self):
+        inner = MockAPIError(
+            "Payment required",
+            status_code=402,
+            body={"error": {"message": "Your credit balance is too low"}},
+        )
+        outer = Exception("outer")
+        outer.__cause__ = inner
+
+        result = classify_api_error(outer)
+
+        assert result.status_code == 402
+        assert result.reason == FailoverReason.billing
+        assert result.retryable is False
+
     def test_disconnect_pattern_ordering(self):
         """Disconnect + large session must beat generic transport catch."""
         class FakeRemoteProtocol(Exception):