Gradescope MCP Server

Instructions

Returns the hierarchical question structure with IDs, types, weights, and
question text. Essential for understanding how an assignment is structured.
Requires instructor/TA access.

Args:
    course_id: The Gradescope course ID.
    assignment_id: The assignment ID.

Implementation Reference

• src/gradescope_mcp/tools/grading.py:132-202 (handler)

  The handler implementation for `tool_get_assignment_outline` which fetches assignment structure and formats it into a markdown outline.

  def get_assignment_outline(course_id: str, assignment_id: str) -> str:
      """Get the question/rubric outline for an assignment.
  
      Returns the hierarchical question structure with IDs, types, weights,
      and question text. This is the foundation for rubric creation and grading.
  
      Args:
          course_id: The Gradescope course ID.
          assignment_id: The assignment ID.
      """
      if not course_id or not assignment_id:
          return "Error: both course_id and assignment_id are required."
  
      try:
          props = _get_outline_data(course_id, assignment_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 fetching outline: {e}"
  
      questions = props.get("questions", {})
      if not questions:
          return f"No questions found for assignment `{assignment_id}`."
  
      tree = _build_question_tree(questions)
  
      # Format output
      assignment_info = props.get("assignment", {})
      lines = [f"## Assignment Outline\n"]
  
      if assignment_info:
          atype = assignment_info.get("type", "Unknown")
          lines.append(f"**Type:** {atype}")
  
      lines.append(f"**Total questions:** {len(questions)}\n")
  
      group_num = 0
      for group in tree:
          group_num += 1
          group_title = group["title"] or f"Question Group {group_num}"
          group_weight = group["weight"]
          lines.append(f"### {group_title} ({group_weight} pts)\n")
  
          if group["children"]:
              lines.append("| # | Question ID | Weight | Type | Question Text |")
              lines.append("|---|-------------|--------|------|---------------|")
              for i, child in enumerate(group["children"], 1):
                  text = _extract_text_content(child["content"])
                  # Truncate for table display
                  if len(text) > 120:
                      text = text[:117] + "..."
                  # Escape pipes in text
                  text = text.replace("|", "\\|")
                  lines.append(
                      f"| {group_num}.{i} | `{child['id']}` | {child['weight']} | "
                      f"{child['type']} | {text} |"
                  )
              lines.append("")
          else:
              # It's a standalone question (no children)
              if group["id"]:
                  lines.append(f"**Question ID:** `{group['id']}`")
              text = _extract_text_content(group["content"])
              if text:
                  lines.append(f"_{text[:200]}_\n")
              else:
                  lines.append("")
  
      return "\n".join(lines)

• src/gradescope_mcp/server.py:288-299 (registration)

  Tool registration for `tool_get_assignment_outline` in the MCP server.

  def tool_get_assignment_outline(course_id: str, assignment_id: str) -> str:
      """Get the question/rubric outline for an assignment.
  
      Returns the hierarchical question structure with IDs, types, weights, and
      question text. Essential for understanding how an assignment is structured.
      Requires instructor/TA access.
  
      Args:
          course_id: The Gradescope course ID.
          assignment_id: The assignment ID.
      """
      return get_assignment_outline(course_id, assignment_id)