Package ai.acolite.agentsdk.core

Interface OutputGuardrail<TContext,TOutput extends AgentOutputType>

public interface OutputGuardrail<TContext,TOutput extends AgentOutputType>

OutputGuardrail

Validates agent output after the final model response.

Output guardrails always run in parallel (no blocking mode) and execute after the agent has produced its final output.

Use output guardrails to: - Check for inappropriate content in responses - Validate output format or structure - Log/audit agent responses - Apply business rules to outputs

Ported from TypeScript OpenAI Agents SDK Source: guardrail.ts

Method Summary

Modifier and Type	Method	Description
CompletableFuture<GuardrailFunctionOutput>	execute(OutputGuardrailFunctionArgs<TContext,TOutput> args)	Execute the guardrail validation on agent output.
String	getName()	Unique name for this guardrail.

Method Details

getName

String getName()

Unique name for this guardrail.

Returns:

The guardrail name

execute

CompletableFuture<GuardrailFunctionOutput> execute(OutputGuardrailFunctionArgs<TContext,TOutput> args)

Execute the guardrail validation on agent output.

Parameters:

args - The output arguments containing final output and context

Returns:

A CompletableFuture with the guardrail result