add flag allow_turn_resume

docs/_static/llama-stack-spec.html

@@ -4287,6 +4287,9 @@
                     },
                     "tool_config": {
                         "$ref": "#/components/schemas/ToolConfig"
+                    },
+                    "allow_turn_resume": {
+                        "type": "boolean"
                     }
                 },
                 "additionalProperties": false,

docs/_static/llama-stack-spec.yaml

@@ -2779,6 +2779,8 @@ components:
             $ref: '#/components/schemas/AgentTool'
         tool_config:
           $ref: '#/components/schemas/ToolConfig'
+        allow_turn_resume:
+          type: boolean
       additionalProperties: false
       required:
         - messages

llama_stack/apis/agents/agents.py

@@ -296,6 +296,9 @@ class AgentTurnCreateRequest(AgentConfigOverridablePerTurn):
     stream: Optional[bool] = False
     tool_config: Optional[ToolConfig] = None
 
+    # TODO (xiyan): used for backward compatibility, update for 0.1.5
+    allow_turn_resume: Optional[bool] = False
+
 
 @json_schema_type
 class AgentTurnResumeRequest(BaseModel):
@@ -352,6 +355,7 @@ class Agents(Protocol):
         documents: Optional[List[Document]] = None,
         toolgroups: Optional[List[AgentToolGroup]] = None,
         tool_config: Optional[ToolConfig] = None,
+        allow_turn_resume: Optional[bool] = False,
     ) -> Union[Turn, AsyncIterator[AgentTurnResponseStreamChunk]]: ...
 
     @webmethod(

llama_stack/providers/inline/agents/meta_reference/agent_instance.py

@@ -227,7 +227,7 @@ class ChatAgent(ShieldRunnerMixin):
             )
             await self.storage.add_turn_to_session(request.session_id, turn)
 
-            if output_message.tool_calls:
+            if output_message.tool_calls and request.allow_turn_resume:
                 chunk = AgentTurnResponseStreamChunk(
                     event=AgentTurnResponseEvent(
                         payload=AgentTurnResponseTurnAwaitingInputPayload(

llama_stack/providers/inline/agents/meta_reference/agents.py

@@ -146,6 +146,7 @@ class MetaReferenceAgentsImpl(Agents):
         documents: Optional[List[Document]] = None,
         stream: Optional[bool] = False,
         tool_config: Optional[ToolConfig] = None,
+        allow_turn_resume: Optional[bool] = False,
     ) -> AsyncGenerator:
         request = AgentTurnCreateRequest(
             agent_id=agent_id,
@@ -155,6 +156,7 @@ class MetaReferenceAgentsImpl(Agents):
             toolgroups=toolgroups,
             documents=documents,
             tool_config=tool_config,
+            allow_turn_resume=allow_turn_resume,
         )
         if stream:
             return self._create_agent_turn_streaming(request)