@radaros/eval

Evaluation framework for testing and scoring RadarOS agent outputs.

Install

npm install @radaros/eval

Quick Start

import { EvalSuite, contains, regexMatch } from "@radaros/eval";

const suite = new EvalSuite({
  name: "basic-tests",
  agent: myAgent,
  cases: [
    { input: "What is 2+2?", expected: "4" },
    { input: "Say hello", expected: "hello" },
  ],
  scorers: [contains(), regexMatch(/\d+/)],
});

const results = await suite.run();

Documentation

Full docs at radaros.mintlify.dev

License

MIT