Merge branch 'main' into add-reasoning-param-to-model-settings

Author: SuveenE

.github/ISSUE_TEMPLATE/feature_request.md

@@ -10,7 +10,7 @@ assignees: ''
 ### Please read this first
 
 - **Have you read the docs?**[Agents SDK docs](https://openai.github.io/openai-agents-python/)
-- **Have you searched for related issues?** Others may have had similar requesrs
+- **Have you searched for related issues?** Others may have had similar requests
 
 ### Describe the feature
 What is the feature you're requesting? How would it work? Please provide examples and details if possible.

.github/ISSUE_TEMPLATE/question.md

@@ -10,7 +10,7 @@ assignees: ''
 ### Please read this first
 
 - **Have you read the docs?**[Agents SDK docs](https://openai.github.io/openai-agents-python/)
-- **Have you searched for related issues?** Others may have had similar requesrs
+- **Have you searched for related issues?** Others may have had similar requests
 
 ### Question
 Describe your question. Provide details if available.

.vscode/settings.json

@@ -0,0 +1,7 @@
+{
+    "python.testing.pytestArgs": [
+        "tests"
+    ],
+    "python.testing.unittestEnabled": false,
+    "python.testing.pytestEnabled": true
+}

docs/running_agents.md

@@ -53,7 +53,7 @@ The `run_config` parameter lets you configure some global settings for the agent
 -   [`handoff_input_filter`][agents.run.RunConfig.handoff_input_filter]: A global input filter to apply to all handoffs, if the handoff doesn't already have one. The input filter allows you to edit the inputs that are sent to the new agent. See the documentation in [`Handoff.input_filter`][agents.handoffs.Handoff.input_filter] for more details.
 -   [`tracing_disabled`][agents.run.RunConfig.tracing_disabled]: Allows you to disable [tracing](tracing.md) for the entire run.
 -   [`trace_include_sensitive_data`][agents.run.RunConfig.trace_include_sensitive_data]: Configures whether traces will include potentially sensitive data, such as LLM and tool call inputs/outputs.
--   [`workflow_name`][agents.run.RunConfig.workflow_name], [`trace_id`][agents.run.RunConfig.trace_id], [`group_id`][agents.run.RunConfig.group_id]: Sets the tracing workflow name, trace ID and trace group ID for the run. We recommend at least setting `workflow_name`. The session ID is an optional field that lets you link traces across multiple runs.
+-   [`workflow_name`][agents.run.RunConfig.workflow_name], [`trace_id`][agents.run.RunConfig.trace_id], [`group_id`][agents.run.RunConfig.group_id]: Sets the tracing workflow name, trace ID and trace group ID for the run. We recommend at least setting `workflow_name`. The group ID is an optional field that lets you link traces across multiple runs.
 -   [`trace_metadata`][agents.run.RunConfig.trace_metadata]: Metadata to include on all traces.
 
 ## Conversations/chat threads

docs/tracing.md

@@ -101,7 +101,8 @@ To customize this default setup, to send traces to alternative or additional bac
 
 -   [Weights & Biases](https://weave-docs.wandb.ai/guides/integrations/openai_agents)
 -   [Arize-Phoenix](https://docs.arize.com/phoenix/tracing/integrations-tracing/openai-agents-sdk)
--   [MLflow](https://mlflow.org/docs/latest/tracing/integrations/openai-agent)
+-   [MLflow (self-hosted/OSS](https://mlflow.org/docs/latest/tracing/integrations/openai-agent)
+-   [MLflow (Databricks hosted](https://docs.databricks.com/aws/en/mlflow/mlflow-tracing#-automatic-tracing)
 -   [Braintrust](https://braintrust.dev/docs/guides/traces/integrations#openai-agents-sdk)
 -   [Pydantic Logfire](https://logfire.pydantic.dev/docs/integrations/llms/openai/#openai-agents)
 -   [AgentOps](https://docs.agentops.ai/v1/integrations/agentssdk)
@@ -112,3 +113,4 @@ To customize this default setup, to send traces to alternative or additional bac
 -   [Comet Opik](https://www.comet.com/docs/opik/tracing/integrations/openai_agents)
 -   [Langfuse](https://langfuse.com/docs/integrations/openaiagentssdk/openai-agents)
 -   [Langtrace](https://docs.langtrace.ai/supported-integrations/llm-frameworks/openai-agents-sdk)
+-   [Okahu-Monocle](https://github.com/monocle2ai/monocle)

examples/financial_research_agent/manager.py

@@ -38,7 +38,7 @@ async def run(self, query: str) -> None:
         with trace("Financial research trace", trace_id=trace_id):
             self.printer.update_item(
                 "trace_id",
-                f"View trace: https://platform.openai.com/traces/{trace_id}",
+                f"View trace: https://platform.openai.com/traces/trace?trace_id={trace_id}",
                 is_done=True,
                 hide_checkmark=True,
             )

examples/mcp/filesystem_example/main.py

@@ -45,7 +45,7 @@ async def main():
     ) as server:
         trace_id = gen_trace_id()
         with trace(workflow_name="MCP Filesystem Example", trace_id=trace_id):
-            print(f"View trace: https://platform.openai.com/traces/{trace_id}\n")
+            print(f"View trace: https://platform.openai.com/traces/trace?trace_id={trace_id}\n")
             await run(server)
 
 

examples/mcp/sse_example/README.md

@@ -5,7 +5,7 @@ This example uses a local SSE server in [server.py](server.py).
 Run the example via:
 
 ```
-uv run python  python examples/mcp/sse_example/main.py
+uv run python examples/mcp/sse_example/main.py
 ```
 
 ## Details

examples/mcp/sse_example/main.py

@@ -46,7 +46,7 @@ async def main():
     ) as server:
         trace_id = gen_trace_id()
         with trace(workflow_name="SSE Example", trace_id=trace_id):
-            print(f"View trace: https://platform.openai.com/traces/{trace_id}\n")
+            print(f"View trace: https://platform.openai.com/traces/trace?trace_id={trace_id}\n")
             await run(server)
 
 

examples/research_bot/manager.py

@@ -23,7 +23,7 @@ async def run(self, query: str) -> None:
         with trace("Research trace", trace_id=trace_id):
             self.printer.update_item(
                 "trace_id",
-                f"View trace: https://platform.openai.com/traces/{trace_id}",
+                f"View trace: https://platform.openai.com/traces/trace?trace_id={trace_id}",
                 is_done=True,
                 hide_checkmark=True,
             )

examples/research_bot/sample_outputs/product_recs.txt

@@ -3,7 +3,7 @@
 $ uv run python -m examples.research_bot.main
 
 What would you like to research? Best surfboards for beginners. I can catch my own waves, but previously used an 11ft board. What should I look for, what are my options? Various budget ranges.
-View trace: https://platform.openai.com/traces/trace_...
+View trace: https://platform.openai.com/traces/trace?trace_id=trace_...
 Starting research...
 ✅ Will perform 15 searches
 ✅ Searching... 15/15 completed

examples/research_bot/sample_outputs/vacation.txt

@@ -2,7 +2,7 @@
 
 $ uv run python -m examples.research_bot.main
 What would you like to research? Caribbean vacation spots in April, optimizing for surfing, hiking and water sports
-View trace: https://platform.openai.com/traces/trace_....
+View trace: https://platform.openai.com/traces/trace?trace_id=trace_....
 Starting research...
 ✅ Will perform 15 searches
 ✅ Searching... 15/15 completed

pyproject.toml

@@ -13,7 +13,7 @@ dependencies = [
     "typing-extensions>=4.12.2, <5",
     "requests>=2.0, <3",
     "types-requests>=2.0, <3",
-    "mcp; python_version >= '3.10'",
+    "mcp>=1.6.0, <2; python_version >= '3.10'",
 ]
 classifiers = [
     "Typing :: Typed",

src/agents/__init__.py

@@ -100,6 +100,7 @@
     transcription_span,
 )
 from .usage import Usage
+from .version import __version__
 
 
 def set_default_openai_key(key: str, use_for_tracing: bool = True) -> None:
@@ -247,4 +248,5 @@ def enable_verbose_stdout_logging():
     "gen_trace_id",
     "gen_span_id",
     "default_tool_error_function",
+    "__version__",
 ]

src/agents/agent.py

@@ -6,7 +6,7 @@
 from dataclasses import dataclass, field
 from typing import TYPE_CHECKING, Any, Callable, Generic, Literal, cast
 
-from typing_extensions import TypeAlias, TypedDict
+from typing_extensions import NotRequired, TypeAlias, TypedDict
 
 from .guardrail import InputGuardrail, OutputGuardrail
 from .handoffs import Handoff
@@ -44,7 +44,7 @@ class ToolsToFinalOutputResult:
     MaybeAwaitable[ToolsToFinalOutputResult],
 ]
 """A function that takes a run context and a list of tool results, and returns a
-`ToolToFinalOutputResult`.
+`ToolsToFinalOutputResult`.
 """
 
 
@@ -53,6 +53,15 @@ class StopAtTools(TypedDict):
     """A list of tool names, any of which will stop the agent from running further."""
 
 
+class MCPConfig(TypedDict):
+    """Configuration for MCP servers."""
+
+    convert_schemas_to_strict: NotRequired[bool]
+    """If True, we will attempt to convert the MCP schemas to strict-mode schemas. This is a
+    best-effort conversion, so some schemas may not be convertible. Defaults to False.
+    """
+
+
 @dataclass
 class Agent(Generic[TContext]):
     """An agent is an AI model configured with instructions, tools, guardrails, handoffs and more.
@@ -119,6 +128,9 @@ class Agent(Generic[TContext]):
     longer needed.
     """
 
+    mcp_config: MCPConfig = field(default_factory=lambda: MCPConfig())
+    """Configuration for MCP servers."""
+
     input_guardrails: list[InputGuardrail[TContext]] = field(default_factory=list)
     """A list of checks that run in parallel to the agent's execution, before generating a
     response. Runs only if the agent is the first agent in the chain.
@@ -224,7 +236,8 @@ async def get_system_prompt(self, run_context: RunContextWrapper[TContext]) -> s
 
     async def get_mcp_tools(self) -> list[Tool]:
         """Fetches the available tools from the MCP servers."""
-        return await MCPUtil.get_all_function_tools(self.mcp_servers)
+        convert_schemas_to_strict = self.mcp_config.get("convert_schemas_to_strict", False)
+        return await MCPUtil.get_all_function_tools(self.mcp_servers, convert_schemas_to_strict)
 
     async def get_all_tools(self) -> list[Tool]:
         """All agent tools, including MCP tools and function tools."""

src/agents/mcp/util.py

@@ -2,6 +2,8 @@
 import json
 from typing import TYPE_CHECKING, Any
 
+from agents.strict_schema import ensure_strict_json_schema
+
 from .. import _debug
 from ..exceptions import AgentsException, ModelBehaviorError, UserError
 from ..logger import logger
@@ -19,12 +21,14 @@ class MCPUtil:
     """Set of utilities for interop between MCP and Agents SDK tools."""
 
     @classmethod
-    async def get_all_function_tools(cls, servers: list["MCPServer"]) -> list[Tool]:
+    async def get_all_function_tools(
+        cls, servers: list["MCPServer"], convert_schemas_to_strict: bool
+    ) -> list[Tool]:
         """Get all function tools from a list of MCP servers."""
         tools = []
         tool_names: set[str] = set()
         for server in servers:
-            server_tools = await cls.get_function_tools(server)
+            server_tools = await cls.get_function_tools(server, convert_schemas_to_strict)
             server_tool_names = {tool.name for tool in server_tools}
             if len(server_tool_names & tool_names) > 0:
                 raise UserError(
@@ -37,25 +41,37 @@ async def get_all_function_tools(cls, servers: list["MCPServer"]) -> list[Tool]:
         return tools
 
     @classmethod
-    async def get_function_tools(cls, server: "MCPServer") -> list[Tool]:
+    async def get_function_tools(
+        cls, server: "MCPServer", convert_schemas_to_strict: bool
+    ) -> list[Tool]:
         """Get all function tools from a single MCP server."""
 
         with mcp_tools_span(server=server.name) as span:
             tools = await server.list_tools()
             span.span_data.result = [tool.name for tool in tools]
 
-        return [cls.to_function_tool(tool, server) for tool in tools]
+        return [cls.to_function_tool(tool, server, convert_schemas_to_strict) for tool in tools]
 
     @classmethod
-    def to_function_tool(cls, tool: "MCPTool", server: "MCPServer") -> FunctionTool:
+    def to_function_tool(
+        cls, tool: "MCPTool", server: "MCPServer", convert_schemas_to_strict: bool
+    ) -> FunctionTool:
         """Convert an MCP tool to an Agents SDK function tool."""
         invoke_func = functools.partial(cls.invoke_mcp_tool, server, tool)
+        schema, is_strict = tool.inputSchema, False
+        if convert_schemas_to_strict:
+            try:
+                schema = ensure_strict_json_schema(schema)
+                is_strict = True
+            except Exception as e:
+                logger.info(f"Error converting MCP schema to strict mode: {e}")
+
         return FunctionTool(
             name=tool.name,
             description=tool.description or "",
-            params_json_schema=tool.inputSchema,
+            params_json_schema=schema,
             on_invoke_tool=invoke_func,
-            strict_json_schema=False,
+            strict_json_schema=is_strict,
         )
 
     @classmethod

src/agents/model_settings.py

@@ -48,6 +48,10 @@ class ModelSettings:
     reasoning effort. For computer_use_preview: Use 'generate_summary' key with values
     'concise' or 'detailed' to get reasoning summaries."""
 
+    metadata: dict[str, str] | None = None
+    """Metadata to include with the model response call."""
+
+
     store: bool | None = None
     """Whether to store the generated model response for later retrieval.
     Defaults to True if not provided."""

src/agents/models/openai_chatcompletions.py

@@ -540,6 +540,7 @@ async def _fetch_response(
             store=store,
             reasoning_effort=self._non_null_or_not_given(reasoning_effort),
             extra_headers=_HEADERS,
+            metadata=model_settings.metadata,
         )
 
         if isinstance(ret, ChatCompletion):
@@ -927,12 +928,13 @@ def ensure_assistant_message() -> ChatCompletionAssistantMessageParam:
             elif func_call := cls.maybe_function_tool_call(item):
                 asst = ensure_assistant_message()
                 tool_calls = list(asst.get("tool_calls", []))
+                arguments = func_call["arguments"] if func_call["arguments"] else "{}"
                 new_tool_call = ChatCompletionMessageToolCallParam(
                     id=func_call["call_id"],
                     type="function",
                     function={
                         "name": func_call["name"],
-                        "arguments": func_call["arguments"],
+                        "arguments": arguments,
                     },
                 )
                 tool_calls.append(new_tool_call)
@@ -975,7 +977,7 @@ def to_openai(cls, tool: Tool) -> ChatCompletionToolParam:
             }
 
         raise UserError(
-            f"Hosted tools are not supported with the ChatCompletions API. FGot tool type: "
+            f"Hosted tools are not supported with the ChatCompletions API. Got tool type: "
             f"{type(tool)}, tool: {tool}"
         )
 

src/agents/models/openai_responses.py

@@ -248,6 +248,7 @@ async def _fetch_response(
             text=response_format,
             store=self._non_null_or_not_given(model_settings.store),
             reasoning=self._non_null_or_not_given(model_settings.reasoning),
+            metadata=model_settings.metadata,
         )
 
     def _get_client(self) -> AsyncOpenAI:

src/agents/tracing/processors.py

@@ -182,7 +182,6 @@ def __init__(
         # Track when we next *must* perform a scheduled export
         self._next_export_time = time.time() + self._schedule_delay
 
-        self._shutdown_event = threading.Event()
         self._worker_thread = threading.Thread(target=self._run, daemon=True)
         self._worker_thread.start()
 

src/agents/tracing/span_data.py

@@ -236,7 +236,7 @@ def export(self) -> dict[str, Any]:
 
 
 class SpeechSpanData(SpanData):
-    __slots__ = ("input", "output", "model", "model_config", "first_byte_at")
+    __slots__ = ("input", "output", "model", "model_config", "first_content_at")
 
     def __init__(
         self,

src/agents/version.py

@@ -1,7 +1,7 @@
 import importlib.metadata
 
 try:
-    __version__ = importlib.metadata.version("agents")
+    __version__ = importlib.metadata.version("openai-agents")
 except importlib.metadata.PackageNotFoundError:
     # Fallback if running from source without being installed
     __version__ = "0.0.0"