Gradescope MCP Server

Instructions

The artifact includes the prompt, rubric, reference-answer notes, page URLs,
crop regions, and a confidence gate for auto-grading.

Args:
    course_id: The Gradescope course ID.
    assignment_id: The assignment ID.
    question_id: The question ID to prepare.
    submission_id: Optional sample submission ID to use for rubric/page context.

Implementation Reference

• src/gradescope_mcp/tools/grading_workflow.py:284-320 (handler)

  The core logic for preparing the grading artifact.

  def prepare_grading_artifact(
      course_id: str,
      assignment_id: str,
      question_id: str,
      submission_id: str | None = None,
  ) -> str:
      """Prepare a cached markdown artifact in /tmp for an assignment question.
  
      The artifact includes question metadata, prompt text when available, rubric,
      a reference answer or fallback draft, and read-strategy notes for agents.
      """
      if not course_id or not assignment_id or not question_id:
          return "Error: course_id, assignment_id, and question_id are required."
  
      try:
          questions = _fetch_assignment_questions(course_id, assignment_id)
          target = questions.get(str(question_id))
          if not target:
              return (
                  f"Error: question `{question_id}` was not found in assignment "
                  f"`{assignment_id}`."
              )
  
          if submission_id is None:
              submission_id = _find_first_submission_id(course_id, question_id)
  
          ctx = _get_grading_context(course_id, question_id, submission_id)
      except AuthError as e:
          return f"Authentication error: {e}"
      except ValueError as e:
          return f"Error: {e}"
      except Exception as e:
          return f"Error preparing grading artifact: {e}"
  
      props = ctx["props"]
      question = props.get("question", {})
      prompt_text, explanation = _extract_outline_prompt_and_reference(

• src/gradescope_mcp/server.py:620-640 (registration)

  MCP tool registration and wrapper for `tool_prepare_grading_artifact`.

  @mcp.tool()
  def tool_prepare_grading_artifact(
      course_id: str,
      assignment_id: str,
      question_id: str,
      submission_id: str | None = None,
  ) -> str:
      """Prepare a cached /tmp markdown artifact for grading a question.
  
      The artifact includes the prompt, rubric, reference-answer notes, page URLs,
      crop regions, and a confidence gate for auto-grading.
  
      Args:
          course_id: The Gradescope course ID.
          assignment_id: The assignment ID.
          question_id: The question ID to prepare.
          submission_id: Optional sample submission ID to use for rubric/page context.
      """
      return prepare_grading_artifact(
          course_id, assignment_id, question_id, submission_id
      )