Configuration
This page covers SDK-wide defaults that you usually set once during application startup, such as the default OpenAI key or client, the default OpenAI API shape, tracing export defaults, and logging behavior.
These defaults still apply to sandbox-based workflows, but sandbox workspaces, sandbox clients, and session reuse are configured separately.
If you need to configure a specific agent or run instead, start with:
- Agents for instructions, tools, output types, handoffs, and guardrails on a plain
Agent. - Running agents for
RunConfig, sessions, and conversation-state options. - Sandbox agents for
SandboxRunConfig, manifests, capabilities, and sandbox-client-specific workspace setup. - Models for model selection and provider configuration.
- Tracing for per-run tracing metadata and custom trace processors.
Configuration objects and dictionaries
Configuration parameters defined by the SDK generally accept either their typed settings object or a dictionary containing the same fields. This applies across agent, run, model, session, sandbox, and voice configuration boundaries whose type annotations include a dictionary. Nested settings types defined by the SDK can also use dictionaries.
from agents import Agent
agent = Agent(
name="Assistant",
model="gpt-5.6-sol",
model_settings={
"reasoning": {"effort": "high"},
"verbosity": "low",
},
)
The SDK normalizes these dictionaries into the corresponding settings objects. Unknown fields in dataclass configuration types defined by the SDK raise TypeError, which helps catch misspelled option names early. Check the parameter's type annotation or API reference to confirm whether a specific boundary accepts a dictionary.
API keys and clients
By default, the SDK uses the OPENAI_API_KEY environment variable for LLM requests and tracing. The key is resolved when the SDK first creates an OpenAI client (lazy initialization), so set the environment variable before your first model call. If you are unable to set that environment variable before your app starts, you can use the set_default_openai_key() function to set the key.
Alternatively, you can also configure an OpenAI client to be used. By default, the SDK creates an AsyncOpenAI instance, using the API key from the environment variable or the default key set above. You can change this by using the set_default_openai_client() function.
from openai import AsyncOpenAI
from agents import set_default_openai_client
custom_client = AsyncOpenAI(base_url="...", api_key="...")
set_default_openai_client(custom_client)
If you prefer environment-based endpoint configuration, the default OpenAI provider also reads OPENAI_BASE_URL. When you enable Responses websocket transport, it also reads OPENAI_WEBSOCKET_BASE_URL for the websocket /responses endpoint.
export OPENAI_BASE_URL="https://your-openai-compatible-endpoint.example/v1"
export OPENAI_WEBSOCKET_BASE_URL="wss://your-openai-compatible-endpoint.example/v1"
Finally, you can also customize the OpenAI API that is used. By default, we use the OpenAI Responses API. You can override this to use the Chat Completions API by using the set_default_openai_api() function.
OpenAI provider defaults
Providers that use the SDK's OpenAI backend also read SDK-wide defaults when they map model-name strings to models. Use set_default_openai_responses_transport() to make OpenAI Responses models use websocket transport by default:
from agents import set_default_openai_responses_transport
set_default_openai_responses_transport("websocket")
This affects OpenAI Responses models that result when the default OpenAI provider resolves a model name. For provider-level setup, connection reuse, keepalive options, and custom websocket endpoints, see Responses WebSocket transport.
If your OpenAI setup expects provider-level agent registration metadata, configure a default harness ID once at startup:
You can also pass the full registration object:
from agents import OpenAIAgentRegistrationConfig, set_default_openai_agent_registration
set_default_openai_agent_registration(
OpenAIAgentRegistrationConfig(harness_id="your-harness-id")
)
If no SDK default is set, providers that use the SDK's OpenAI backend fall back to the OPENAI_AGENT_HARNESS_ID environment variable. When a harness ID is configured, the SDK adds it to trace metadata as agent_harness_id unless that key is already present in RunConfig.trace_metadata.
Tracing
Tracing is enabled by default. By default it uses the same OpenAI API key as your model requests from the section above (that is, the environment variable or the default key you set). You can specifically set the API key used for tracing by using the set_tracing_export_api_key function.
If your model traffic uses one key or client but tracing should use a different OpenAI key, pass use_for_tracing=False when setting the default key or client, then configure tracing separately. The same pattern works with set_default_openai_key() if you are not using a custom client.
from openai import AsyncOpenAI
from agents import (
set_default_openai_client,
set_tracing_export_api_key,
)
custom_client = AsyncOpenAI(base_url="https://your-openai-compatible-endpoint.example/v1", api_key="provider-key")
set_default_openai_client(custom_client, use_for_tracing=False)
set_tracing_export_api_key("sk-tracing")
If you need to attribute traces to a specific organization or project when using the default exporter, set these environment variables before your app starts:
You can also set a tracing API key per run without changing the global exporter.
from agents import Runner, RunConfig
await Runner.run(
agent,
input="Hello",
run_config=RunConfig(tracing={"api_key": "sk-tracing-123"}),
)
You can also disable tracing entirely by using the set_tracing_disabled() function.
If you want to keep tracing enabled but exclude potentially sensitive inputs/outputs from trace payloads, set RunConfig.trace_include_sensitive_data to False:
from agents import Runner, RunConfig
await Runner.run(
agent,
input="Hello",
run_config=RunConfig(trace_include_sensitive_data=False),
)
You can also change the default without code by setting this environment variable before your app starts:
For full tracing controls, see the tracing guide.
Debug logging
The SDK defines two Python loggers (openai.agents and openai.agents.tracing) and does not attach handlers by default. Logs follow your application's Python logging configuration.
To enable verbose logging, use the enable_verbose_stdout_logging() function.
Alternatively, you can customize the logs by adding handlers, filters, formatters, etc. You can read more in the Python logging guide.
import logging
logger = logging.getLogger("openai.agents") # or openai.agents.tracing for the Tracing logger
# To make all logs show up
logger.setLevel(logging.DEBUG)
# To make info and above show up
logger.setLevel(logging.INFO)
# To make warning and above show up
logger.setLevel(logging.WARNING)
# etc
# You can customize this as needed, but this will output to `stderr` by default
logger.addHandler(logging.StreamHandler())
Sensitive data in logs and diagnostics
Certain logs and diagnostic exceptions may contain sensitive data (for example, model or tool inputs and outputs).
By default, the SDK does not log LLM inputs/outputs or tool inputs/outputs. These protections are controlled by:
If you need to include this data temporarily for debugging, set either variable to 0 (or false) before your app starts:
These flags also control whether affected failures retain payload-bearing diagnostic details. For example, with tool-data redaction enabled, invalid arguments for a FunctionTool raise a generic ModelBehaviorError without chaining the underlying validation error. Setting either variable to 0 can expose raw model or tool data in logs, exception messages, exception chains, and other diagnostic context, so enable it only in a controlled development environment.