Disable extended thinking: causes blank responses via streaming retry

agentclaw/app/agent_claw_main/main.py

@@ -272,15 +272,13 @@ async def main(payload: dict, context):
     system_prompt = system_prompt + f'\n\nCurrent date/time: {_time_str}'
     print(f'[main] System prompt time injection: {_time_str}')
 
-    # Model: claude-sonnet-4-6 via cross-region inference with extended thinking
+    # Model: claude-sonnet-4-6 via cross-region inference
+    # NOTE: extended thinking disabled — causes retry/duplicate issues with streaming
     from botocore.config import Config as BotoConfig
     model = BedrockModel(
         model_id="us.anthropic.claude-sonnet-4-6",
         region_name="us-east-1",
         boto_client_config=BotoConfig(read_timeout=600, connect_timeout=10),
-        additional_model_request_fields={
-            "thinking": {"type": "enabled", "budget_tokens": 2000}
-        },
     )
 
     base_tools = [web_search, web_fetch, read_workspace_file, write_workspace_file,