Skip to content
Docs menu

Tool reference · Team onboarding

New in 14.6.0

onboarding_grade

Grades or regrades an open answer from the grading queue, with points and an optional comment for the employee.

team server /mcp
Caution. Only open answers can be graded, and the score cannot exceed the question's points. Requires the department head role or superadmin.

When to use

  • An attempt is waiting for review because the server has no LLM or the LLM grader failed.
  • The department head disagrees with an LLM grade and wants to regrade.

Parameters

NameTypeDefaultDescription
attempt_idrequired integer — —
question_idrequired string — (pattern ^[a-zA-Z0-9_-]{1,64}$)
scorerequired number — Points earned, from 0 to the question's points. (min 0, max 100)
comment string "" (max length 2,000)

Example

Arguments

{
  "attempt_id": 17,
  "question_id": "q_billing",
  "score": 2,
  "comment": "Right: the second reviewer is the point."
}

Result shape

{
  "attempt_id": 17,
  "module_id": "mod_deploy",
  "status": "graded",
  "score": 3,
  "max_score": 3,
  "percent": 100,
  "pass_threshold": 0.8,
  "passed": true,
  "attempts_used": 1,
  "attempts_left": 2,
  "answers_revealed": true,
  "results": [
    {
      "question_id": "q_billing",
      "status": "graded",
      "earned": 2,
      "points": 2,
      "grader": "manager",
      "comment": "Right: the second reviewer is the point.",
      "correct": true,
      "correct_options": null,
      "rubric": "Billing is high-risk; a second reviewer catches mistakes."
    }
  ]
}

Grading the last pending answer finalises the attempt; regrading a finalised answer recomputes the result.

Values are illustrative; the keys follow the server's handler. MCP clients receive the result as JSON text content.

Server description

The description the server sends to your agent in tools/list, captured from the v14.7.0 source:

Department head: grade or regrade an open answer (points 0..question points) from the grading queue in onboarding_team_report.

Found a mistake? Open an issue on GitHub.

Search