fix(errors): classify Z.AI GLM token-limit message as context overflow (port opencode#35671)

agent/error_classifier.py

@@ -267,6 +267,8 @@ def is_auth(self) -> bool:
     # Chinese error messages (some providers return these)
     "超过最大长度",
     "上下文长度",
+    # Z.AI / Zhipu GLM pattern (English form; error code 1210)
+    "tokens in request more than max tokens allowed",
     # AWS Bedrock Converse API error patterns
     "input is too long",
     "max input token",

tests/agent/test_error_classifier.py

@@ -1226,6 +1226,20 @@ def test_chinese_context_overflow(self):
         result = classify_api_error(e)
         assert result.reason == FailoverReason.context_overflow
 
+    # ── Z.AI / Zhipu GLM error messages ──
+
+    def test_zai_glm_token_limit_overflow(self):
+        """Z.AI GLM's 'tokens in request more than max tokens allowed'
+        (error code 1210) → context_overflow, so the agent compresses
+        instead of blindly retrying. Port of anomalyco/opencode#35671."""
+        e = MockAPIError(
+            '{"error": {"code": "1210", "message": '
+            '"tokens in request more than max tokens allowed"}}',
+            status_code=400,
+        )
+        result = classify_api_error(e, provider="zai")
+        assert result.reason == FailoverReason.context_overflow
+
     # ── vLLM / local inference server error messages ──
 
     def test_vllm_max_model_len_overflow(self):