fix: stream never ruff

api/core/agent/base_agent_runner.py

@@ -253,9 +253,11 @@ class BaseAgentRunner(AppRunner):
             # save tool entity
             tool_instances[dataset_tool.entity.identity.name] = dataset_tool
 
-        output_tools = build_agent_output_tools(tenant_id=self.tenant_id,
-                                                invoke_from=self.application_generate_entity.invoke_from,
-                                                tool_invoke_from=ToolInvokeFrom.AGENT)
+        output_tools = build_agent_output_tools(
+            tenant_id=self.tenant_id,
+            invoke_from=self.application_generate_entity.invoke_from,
+            tool_invoke_from=ToolInvokeFrom.AGENT,
+        )
         for tool in output_tools:
             tool_instances[tool.entity.identity.name] = tool
 

api/core/agent/entities.py

@@ -194,6 +194,7 @@ class AgentResult(BaseModel):
     """
     Agent execution result.
     """
+
     output: str | dict = Field(default="", description="The generated output")
     files: list[Any] = Field(default_factory=list, description="Files produced during execution")
     usage: Any | None = Field(default=None, description="LLM usage statistics")

api/core/agent/output_tools.py

@@ -101,6 +101,7 @@ def build_agent_output_tools(
             message_id: str | None = None,
         ) -> ToolInvokeMessage:
             return ToolInvokeMessage(message=ToolInvokeMessage.TextMessage(text=TERMINAL_OUTPUT_MESSAGE))
+
         raw_tool._invoke = invoke_tool  # pyright: ignore[reportPrivateUsage]
         tools.append(raw_tool)
 

api/core/agent/patterns/base.py

@@ -57,10 +57,7 @@ class AgentPattern(ABC):
 
     @abstractmethod
     def run(
-        self,
-        prompt_messages: list[PromptMessage],
-        model_parameters: dict[str, Any],
-        stop: list[str]
+        self, prompt_messages: list[PromptMessage], model_parameters: dict[str, Any], stop: list[str]
     ) -> Generator[LLMResultChunk | AgentLog, None, AgentResult]:
         """Execute the agent strategy."""
         pass

api/core/agent/patterns/function_call.py

@@ -30,10 +30,7 @@ class FunctionCallStrategy(AgentPattern):
     """Function Call strategy using model's native tool calling capability."""
 
     def run(
-        self,
-        prompt_messages: list[PromptMessage],
-        model_parameters: dict[str, Any],
-        stop: list[str]
+        self, prompt_messages: list[PromptMessage], model_parameters: dict[str, Any], stop: list[str]
     ) -> Generator[LLMResultChunk | AgentLog, None, AgentResult]:
         """Execute the function call agent strategy."""
         # Convert tools to prompt format
@@ -144,9 +141,7 @@ class FunctionCallStrategy(AgentPattern):
                 output_files.extend(tool_files)
                 if tool_response == TERMINAL_OUTPUT_MESSAGE:
                     function_call_state = False
-                    final_tool_args = tool_entity.transform_tool_parameters_type(
-                        tool_args
-                    )
+                    final_tool_args = tool_entity.transform_tool_parameters_type(tool_args)
 
             yield self._finish_log(
                 round_log,

api/core/agent/patterns/react.py

@@ -59,11 +59,7 @@ class ReActStrategy(AgentPattern):
         self.instruction = instruction
 
     def run(
-        self,
-        prompt_messages:
-        list[PromptMessage],
-        model_parameters: dict[str, Any],
-        stop: list[str]
+        self, prompt_messages: list[PromptMessage], model_parameters: dict[str, Any], stop: list[str]
     ) -> Generator[LLMResultChunk | AgentLog, None, AgentResult]:
         """Execute the ReAct agent strategy."""
         # Initialize tracking

api/core/agent/patterns/strategy_factory.py

@@ -43,7 +43,7 @@ class StrategyFactory:
         agent_strategy: AgentEntity.Strategy | None = None,
         tool_invoke_hook: ToolInvokeHook | None = None,
         instruction: str = "",
-        structured_output_schema: Mapping[str, Any] | None = None
+        structured_output_schema: Mapping[str, Any] | None = None,
     ) -> AgentPattern:
         """
         Create an appropriate strategy based on model features.
@@ -71,7 +71,7 @@ class StrategyFactory:
             tenant_id=tenant_id,
             invoke_from=invoke_from,
             tool_invoke_from=tool_invoke_from,
-            structured_output_schema=structured_output_schema
+            structured_output_schema=structured_output_schema,
         )
 
         tools.extend(output_tools)

api/core/llm_generator/llm_generator.py

@@ -563,7 +563,7 @@ class LLMGenerator:
                 model_instance=model_instance,
                 prompt_messages=prompt_messages,
                 output_model=SuggestedQuestionsOutput,
-                model_parameters=completion_params
+                model_parameters=completion_params,
             )
 
             return {"questions": response.questions, "error": ""}
@@ -849,11 +849,14 @@ Generate {language} code to extract/transform available variables for the target
         try:
             from core.llm_generator.output_parser.structured_output import invoke_llm_with_pydantic_model
 
-            response = invoke_llm_with_pydantic_model(provider=model_instance.provider, model_schema=model_schema,
-                                                      model_instance=model_instance,
-                                                      prompt_messages=list(prompt_messages),
-                                                      output_model=InstructionModifyOutput,
-                                                      model_parameters=model_parameters)
+            response = invoke_llm_with_pydantic_model(
+                provider=model_instance.provider,
+                model_schema=model_schema,
+                model_instance=model_instance,
+                prompt_messages=list(prompt_messages),
+                output_model=InstructionModifyOutput,
+                model_parameters=model_parameters,
+            )
             return response.model_dump(mode="python")
         except InvokeError as e:
             error = str(e)

api/core/plugin/backwards_invocation/model.py

@@ -117,7 +117,7 @@ class PluginModelBackwardsInvocation(BaseBackwardsInvocation):
             model_parameters=payload.completion_params,
             tools=payload.tools,
             stop=payload.stop,
-            user=user_id
+            user=user_id,
         )
 
         if response.usage:

api/services/workflow_comment_service.py

@@ -69,9 +69,7 @@ class WorkflowCommentService:
         if not candidate_user_ids:
             return []
 
-        app_name = session.scalar(
-            select(App.name).where(App.id == app_id, App.tenant_id == tenant_id)
-        ) or "Dify app"
+        app_name = session.scalar(select(App.name).where(App.id == app_id, App.tenant_id == tenant_id)) or "Dify app"
         commenter_name = session.scalar(select(Account.name).where(Account.id == mentioner_id)) or "Dify user"
         comment_excerpt = WorkflowCommentService._format_comment_excerpt(content)
 
@@ -368,9 +366,7 @@ class WorkflowCommentService:
             mentioned_user_ids = WorkflowCommentService._filter_valid_mentioned_user_ids(mentioned_user_ids or [])
             for user_id in mentioned_user_ids:
                 # Create mention linking to specific reply
-                mention = WorkflowCommentMention(
-                    comment_id=comment_id, reply_id=reply.id, mentioned_user_id=user_id
-                )
+                mention = WorkflowCommentMention(comment_id=comment_id, reply_id=reply.id, mentioned_user_id=user_id)
                 session.add(mention)
 
             mention_email_payloads = WorkflowCommentService._build_mention_email_payloads(