Package com.tnsai.coordination.judge

Interface JudgePolicy

All Known Implementing Classes:

LLMJudgePolicy, ThresholdJudgePolicy

Functional Interface:

This is a functional interface and can therefore be used as the assignment target for a lambda expression or method reference.

@FunctionalInterface
public interface JudgePolicy

Strategy interface for evaluating coordination cycle output.

Implementations decide whether an agent's output meets quality criteria. The JudgeCoordinator calls this at the end of each cycle to determine whether to approve, reject, or retry.

Two built-in implementations are provided:

• ThresholdJudgePolicy — simple numeric threshold check
• LLMJudgePolicy — LLM-based evaluation with natural language criteria

Since:

0.2.3

See Also:

• JudgeCoordinator

Method Summary

Modifier and Type	Method	Description
JudgeVerdict	evaluate(String agentId, Object output, JudgeContext context)	Evaluates an agent's output and produces a verdict.

Method Details

evaluate

JudgeVerdict evaluate(String agentId,
 Object output,
 JudgeContext context)

Evaluates an agent's output and produces a verdict.

Parameters:

agentId - the identifier of the agent whose output is being judged

output - the output to evaluate (type depends on coordination pattern)

context - iteration state and history

Returns:

a verdict with decision, score, and optional feedback