Verify: new-hire onboarding completion

The job

Jev judge eval (2 boolean) over a fixed state : runnable via experimental_evaluate.

The prompt

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

// Verify: new-hire onboarding completion — Jev judge eval (typesafe-ai/jev).
// An agent reports onboarding done; the judge verifies access and equipment claims.
// 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: "Agent report: Sara (starts Monday) — laptop ordered (arrives Friday), SSO + GitHub + Slack provisioned, day-1 buddy assigned (Tom), payroll invite sent. Pending: desk assignment from facilities.",
  questions: {
    "access_ready": {
      "type": "boolean",
      "instructions": "True only if all day-1 system access (SSO, GitHub, Slack or equivalents) is confirmed provisioned."
    },
    "equipment_ready": {
      "type": "boolean",
      "instructions": "True only if the laptop or equivalent equipment is confirmed ordered with an arrival date."
    }
  },
});

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