Hiring

Mark written answers

hiring_mark_assessment · Changes data · Markdown

Gives points to the written answers of a completed test.

$0.01 a call

From your agent’s balance. A call that fails costs nothing.

An example

Give Maria 4 of 5 points for her answer about the creak.

Your agent callshiring_mark_assessment$0.01 a call

The example business is Halverson Cycle & Repair, a bike shop with a repair workshop in Portland, Oregon. The reply is the one the tool really gives.

With these arguments
{
  "applicant_id": 4101,
  "assessment": "workshop-basics",
  "marks": [
    {
      "question": "q2",
      "points": 4
    }
  ]
}
What comes back
{
  "result": {
    "assessment": "workshop-basics",
    "title": "Workshop basics",
    "status": "completed",
    "sent_at": "2026-10-12T09:15:00+00:00",
    "link": null,
    "expires_at": null,
    "score": 5,
    "max_score": 6,
    "written_to_mark": 0,
    "questions": [
      {
        "id": "q1",
        "type": "choice",
        "prompt": "A carbon seatpost clamp is marked 5 Nm. How do you tighten it?",
        "max_points": 1,
        "answer": "To 5 Nm with a torque wrench",
        "expected": [
          "To 5 Nm with a torque wrench"
        ],
        "points": 1
      },
      {
        "id": "q2",
        "type": "long_text",
        "prompt": "A customer's bike creaks under load, but only some days. How would you find the cause?",
        "max_points": 5,
        "answer": "Ask when it happens, then rule out the cheap things in order: pedals and cleats, the saddle rails, the seatpost, then the bottom bracket. Grease and torque each one and test ride between steps.",
        "points": 4,
        "to_mark": false
      }
    ],
    "away_from_page": 0,
    "minutes_taken": 14.5,
    "completed_at": "2026-10-13T16:40:00+00:00"
  }
}

Use it when

A completed test has written answers waiting for marks.

What it does

Give points to the written answers of a completed test (0 up to each question's points).

What comes back

result: the test with its new score. Every reply also carries cost_usd and balance_usd.

Inputs

NameTypeWhat it is
applicant_idRequiredintegerAt least 1.
assessmentRequiredstringThe test (its slug from hiring_list_assessments). 1 to 80 characters.
marksRequiredarray of objects[{"question": "q4", "points": 3}]. 1 to 50 items.

The call

Over MCP, call hiring_mark_assessment on https://api.conecto.chat/mcp/hiring. Over REST:

Terminal
curl -X POST https://api.conecto.chat/v1/tools/hiring_mark_assessment \
  -H "Authorization: Bearer $CONECTO_KEY" \
  -H "Content-Type: application/json" \
  -d '{"applicant_id":4101,"assessment":"workshop-basics","marks":[{"question":"q2","points":4}]}'

When it refuses

A refused call costs nothing and says what to do next.

CodeWhenWhat to do
not_found404No applicant of the business has that applicant_id.Find them with hiring_list_applicants.
not_found404No skills test has that slug.Call hiring_list_assessments and use one of its slugs.
not_completed409The applicant has not completed the test.Wait for hiring.test_completed.
payment_required402The test was completed while the balance could not pay its fee.Top up; the result opens once its fee is paid.
invalid_input400A mark names a question that is not a written answer, or has too many points.Mark only written answers, from 0 to each question's points.

Give this to your agent.

Set up Conecto from https://conecto.chat/SKILL.md