fix(advisor): inject max_uses_exceeded error result instead of raising exception

When the advisor loop hits max_uses, inject a tool_result error so the executor
sees the cap and continues without further advice — matches Anthropic server-side
behaviour (error_code: max_uses_exceeded).

litellm/llms/anthropic/experimental_pass_through/messages/interceptors/advisor.py

@@ -31,10 +31,6 @@ ADVISOR_TOOL_DESCRIPTION: str = _c.ADVISOR_TOOL_DESCRIPTION
 from .base import MessagesInterceptor
 
 
-class AdvisorMaxIterationsError(Exception):
-    """Raised when the advisor loop exceeds max_uses."""
-
-
 class AdvisorOrchestrationHandler(MessagesInterceptor):
     """Orchestrates the advisor tool loop for non-native providers."""
 
@@ -135,10 +131,12 @@ class AdvisorOrchestrationHandler(MessagesInterceptor):
 
             iteration += 1
             if iteration > max_uses:
-                raise AdvisorMaxIterationsError(
-                    f"Advisor orchestration loop exceeded max_uses={max_uses}. "
-                    "Increase max_uses in the advisor tool definition or cap the request."
+                # Per Anthropic spec: inject max_uses_exceeded error result so the
+                # executor sees the cap and continues without further advice.
+                current_messages = _inject_max_uses_error(
+                    current_messages, executor_response, advisor_use_block
                 )
+                continue
 
             # --- Build advisor context ---
             advisor_messages = _build_advisor_context(
@@ -290,6 +288,35 @@ def _inject_advisor_turn(
     ]
 
 
+def _inject_max_uses_error(
+    messages: List[Dict],
+    executor_response: Any,
+    advisor_use_block: Dict,
+) -> List[Dict]:
+    """
+    Inject a max_uses_exceeded error tool_result so the executor continues
+    without further advisor calls (mirrors Anthropic's server-side behaviour).
+    """
+    executor_content = (
+        executor_response.get("content") if isinstance(executor_response, dict) else []
+    ) or []
+    tool_use_id = advisor_use_block.get("id", "")
+    return [
+        *messages,
+        {"role": "assistant", "content": executor_content},
+        {
+            "role": "user",
+            "content": [
+                {
+                    "type": "tool_result",
+                    "tool_use_id": tool_use_id,
+                    "content": "Advisor unavailable: max_uses limit reached. Continue without advisor guidance.",
+                }
+            ],
+        },
+    ]
+
+
 async def _call_messages_handler(
     model: str,
     messages: List[Dict],