Heppu APIv2.heppu.ai/api

List workers

The Workforce page, as JSON.

GET/api/v2/agentskey or session

Returns every worker in the org with workforce fields. No filters, no pagination, ordered newest first. Most orgs have a handful of workers, so this is fine.

curl https://v2.heppu.ai/api/v2/agents -H "x-api-key: hep_..."
200
{
  "data": {
    "agents": [
      {
        "id": "6f0c1c0a-1d9d-4a1e-9c53-2b2a1b0f9c11",
        "name": "Sales Voice",
        "type": "voice",
        "status": "active",
        "model": "gpt-realtime-2.1",
        "specialty": "Outbound sales",
        "displayRole": "Voice employee",
        "workforceStatus": "active",
        "totalActions": 128,
        "successfulActions": 121,
        "reliabilityScore": "0.9453",
        "phoneNumbers": ["+358401234567"],
        "assignedJobs": [
          { "id": "…", "name": "Q3 outreach", "status": "running", "triggerType": "schedule", "trustLevel": "supervised" }
        ],
        "createdAt": "2026-09-01T08:15:00.000Z",
        "updatedAt": "2026-09-10T12:00:00.000Z"
      }
    ],
    "count": 1
  },
  "meta": { "timestamp": "2026-09-11T10:00:00.000Z" }
}

Things to know:

  • reliabilityScore is a string, 0 to 1, four decimals. Parse it.
  • workforceStatus is active, idle or inactive. Separate from status.
  • phoneNumbers is always there, empty for chat workers.
  • assignedJobs[].trustLevel is supervised, semi_autonomous or autonomous.
  • No instructions, config or tools here. Use GET /v1/agents/{id} for those.

Needs the agent:read capability. Every role has it.