GPT-5, GPT-5-mini support

Improvements to model name resolution
Improved instructions for multi-step workflows when continuation is available
Improved instructions for chat tool
Improved preferred model resolution, moved code from registry -> each provider
Updated tests

tools/workflow/base.py

@@ -390,6 +390,23 @@ class WorkflowTool(BaseTool, BaseWorkflowMixin):
         """Get status for skipped expert analysis. Override for tool-specific status."""
         return "skipped_by_tool_design"
 
+    def is_continuation_workflow(self, request) -> bool:
+        """
+        Check if this is a continuation workflow that should skip multi-step investigation.
+
+        When continuation_id is provided, the workflow typically continues from a previous
+        conversation and should go directly to expert analysis rather than starting a new
+        multi-step investigation.
+
+        Args:
+            request: The workflow request object
+
+        Returns:
+            True if this is a continuation that should skip multi-step workflow
+        """
+        continuation_id = self.get_request_continuation_id(request)
+        return bool(continuation_id)
+
     # Abstract methods that must be implemented by specific workflow tools
     # (These are inherited from BaseWorkflowMixin and must be implemented)
 

tools/workflow/workflow_mixin.py

@@ -663,13 +663,13 @@ class BaseWorkflowMixin(ABC):
                 self._current_model_name = None
                 self._model_context = None
 
+            # Handle continuation
+            continuation_id = request.continuation_id
+
             # Adjust total steps if needed
             if request.step_number > request.total_steps:
                 request.total_steps = request.step_number
 
-            # Handle continuation
-            continuation_id = request.continuation_id
-
             # Create thread for first step
             if not continuation_id and request.step_number == 1:
                 clean_args = {k: v for k, v in arguments.items() if k not in ["_model_context", "_resolved_model_name"]}