fix: agent config validation (#1053)

Summary:

Fixes AgentConfig init bug introduced with ToolConfig.

Namely, the below doesn't work
```
    agent_config = AgentConfig(
        **common_params,
        tool_config=ToolConfig(
            tool_choice="required",
        ),
    )
```
bvecause tool_choice was defaulted to 'auto' leading to validation check
failing.

Test Plan:

added unittests

LLAMA_STACK_CONFIG=fireworks pytest -s -v tests/client-sdk/
--safety-shield meta-llama/Llama-Guard-3-8B

llama_stack/apis/agents/agents.py

@@ -154,7 +154,7 @@ class AgentConfigCommon(BaseModel):
     output_shields: Optional[List[str]] = Field(default_factory=list)
     toolgroups: Optional[List[AgentToolGroup]] = Field(default_factory=list)
     client_tools: Optional[List[ToolDef]] = Field(default_factory=list)
-    tool_choice: Optional[ToolChoice] = Field(default=ToolChoice.auto, deprecated="use tool_config instead")
+    tool_choice: Optional[ToolChoice] = Field(default=None, deprecated="use tool_config instead")
     tool_prompt_format: Optional[ToolPromptFormat] = Field(default=None, deprecated="use tool_config instead")
     tool_config: Optional[ToolConfig] = Field(default=None)
 
@@ -166,11 +166,13 @@ class AgentConfigCommon(BaseModel):
                 raise ValueError("tool_choice is deprecated. Use tool_choice in tool_config instead.")
             if self.tool_prompt_format and self.tool_config.tool_prompt_format != self.tool_prompt_format:
                 raise ValueError("tool_prompt_format is deprecated. Use tool_prompt_format in tool_config instead.")
-        if self.tool_config is None:
+        else:
-            self.tool_config = ToolConfig(
+            params = {}
-                tool_choice=self.tool_choice,
+            if self.tool_choice:
-                tool_prompt_format=self.tool_prompt_format,
+                params["tool_choice"] = self.tool_choice
-            )
+            if self.tool_prompt_format:
+                params["tool_prompt_format"] = self.tool_prompt_format
+            self.tool_config = ToolConfig(**params)
 
 
 @json_schema_type

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

@@ -81,12 +81,6 @@ class MetaReferenceAgentsImpl(Agents):
     ) -> AgentCreateResponse:
         agent_id = str(uuid.uuid4())
 
-        if agent_config.tool_config is None:
-            agent_config.tool_config = ToolConfig(
-                tool_choice=agent_config.tool_choice,
-                tool_prompt_format=agent_config.tool_prompt_format,
-            )
-
         await self.persistence_store.set(
             key=f"agent:{agent_id}",
             value=agent_config.model_dump_json(),

tests/client-sdk/agents/test_agents.py

@@ -13,11 +13,12 @@ from llama_stack_client.lib.agents.agent import Agent
 from llama_stack_client.lib.agents.client_tool import ClientTool
 from llama_stack_client.lib.agents.event_logger import EventLogger
 from llama_stack_client.types import ToolResponseMessage
-from llama_stack_client.types.agent_create_params import AgentConfig
+from llama_stack_client.types.shared_params.agent_config import AgentConfig, ToolConfig
 from llama_stack_client.types.agents.turn_create_params import Document as AgentDocument
 from llama_stack_client.types.memory_insert_params import Document
 from llama_stack_client.types.shared.completion_message import CompletionMessage
 from llama_stack_client.types.tool_def_param import Parameter
+from llama_stack.apis.agents.agents import AgentConfig as Server__AgentConfig, ToolChoice
 
 
 class TestClientTool(ClientTool):
@@ -141,6 +142,62 @@ def test_agent_simple(llama_stack_client, agent_config):
         assert "I can't" in logs_str
 
 
+def test_tool_config(llama_stack_client, agent_config):
+    common_params = dict(
+        model="meta-llama/Llama-3.2-3B-Instruct",
+        instructions="You are a helpful assistant",
+        sampling_params={
+            "strategy": {
+                "type": "top_p",
+                "temperature": 1.0,
+                "top_p": 0.9,
+            },
+        },
+        toolgroups=[],
+        enable_session_persistence=False,
+    )
+    agent_config = AgentConfig(
+        **common_params,
+    )
+    Server__AgentConfig(**agent_config)
+
+    agent_config = AgentConfig(
+        **common_params,
+        tool_choice="auto",
+    )
+    server_config = Server__AgentConfig(**agent_config)
+    assert server_config.tool_config.tool_choice == ToolChoice.auto
+
+    agent_config = AgentConfig(
+        **common_params,
+        tool_choice="auto",
+        tool_config=ToolConfig(
+            tool_choice="auto",
+        ),
+    )
+    server_config = Server__AgentConfig(**agent_config)
+    assert server_config.tool_config.tool_choice == ToolChoice.auto
+
+    agent_config = AgentConfig(
+        **common_params,
+        tool_config=ToolConfig(
+            tool_choice="required",
+        ),
+    )
+    server_config = Server__AgentConfig(**agent_config)
+    assert server_config.tool_config.tool_choice == ToolChoice.required
+
+    agent_config = AgentConfig(
+        **common_params,
+        tool_choice="required",
+        tool_config=ToolConfig(
+            tool_choice="auto",
+        ),
+    )
+    with pytest.raises(ValueError, match="tool_choice is deprecated"):
+        Server__AgentConfig(**agent_config)
+
+
 def test_builtin_tool_web_search(llama_stack_client, agent_config):
     agent_config = {
         **agent_config,