OpenAI Agents SDK

pip install "guardlayer[openai-agents]"

from agents import Agent, Runner, function_tool
from guardlayer import GuardLayer
from guardlayer.integrations.openai_agents import guardrails

gl = guardrails(GuardLayer())
agent = Agent(
    name="assistant",
    input_guardrails=[gl.input],
    output_guardrails=[gl.output],
    tools=[function_tool(fetch, tool_input_guardrails=[gl.tool_input], tool_output_guardrails=[gl.tool_output])],
)
await Runner.run(agent, prompt, context={"session_id": "user-42"})

Guardrail	Behaviour
input / output	Trip the SDK's tripwire (InputGuardrailTripwireTriggered / OutputGuardrailTripwireTriggered) when GuardLayer doesn't allow the text.
tool_input	A blocked or review-level call is rejected; the model gets a refusal instead of a result.
tool_output	Output containing an injection is replaced by a notice.

The session is read from the run context (context={"session_id": ...}, or an object with a session_id attribute). Pass guardrails(guard, session=...) to override it. Without a session id, each check is stateless.

Human approval

GuardLayer can't pause an Agents SDK run by itself. For real human approval, also set the SDK's needs_approval on the tool; GuardLayer's review verdict then refuses rather than waits.