Evals · JSON Structure
Evals Judge Structure
A specialized LLM-as-a-judge (or human) scorer with model, prompt, rubric, and calibration data.
EvalsLLM EvaluationAI QualityBenchmarksLLM as a JudgeObservabilityAgent EvaluationRAG EvaluationTest-Driven AI
Judge is a JSON Structure definition published by Evals, describing 10 properties, of which 3 are required. It conforms to the https://json-structure.org/meta/core/v0/# meta-schema.
Properties
id
name
description
judge_kind
model
prompt_template
rubric
output_format
calibration
tags
Meta-schema: https://json-structure.org/meta/core/v0/#