Grade: meeting notes usefulness

The job

Jev judge eval (1 score) over a fixed state : runnable via experimental_evaluate.

The prompt

Copy this into Muse and replace anything in [square brackets] with your own details.

// Grade: meeting notes usefulness — Jev judge eval (typesafe-ai/jev).
// An agent turns a rambling sync into notes; the judge scores usefulness.
// Model: typesafe-ai/jev via the Vercel AI Gateway ($0.042/1M input tokens).
// Auth: set AI_GATEWAY_API_KEY (vck_…) in the environment — never paste the key here.
// For zero-data-retention / no-training workloads, add
// providerOptions: { gateway: { … } } to the evaluate call below.

import { experimental_evaluate } from 'ai';

const result = await experimental_evaluate({
  model: 'typesafe-ai/jev',
  state: "Notes: Decision — ship the CSV export first, SSO second (owner: Priya, dates: Mar 4 / Mar 18). Open question: pricing for exports over 1M rows (owner: Sam, due Feb 27). Dropped: the PDF redesign — revisit in Q3.",
  questions: {
    "notes": {
      "type": "score",
      "instructions": "Score how useful these meeting notes are. Higher is better.",
      "criteria": [
        "Raw transcript with no decisions or owners.",
        "Captures discussion but decisions and owners are unclear.",
        "Decisions clear, but owners or dates missing.",
        "Decisions, owners, dates, open questions, and explicit drops."
      ]
    }
  },
});

console.log(JSON.stringify(result, null, 2));