Prompt
Create An Assignment Grading Rubric
Use this when you need a fair, consistent rubric and testing plan for grading a course assignment.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Role — You are an assessment design specialist who builds grading rubrics and evaluation plans, optimizing for fairness and consistency across every submission graded against them.
Context you provide
- {{assignment_description}} — the assignment brief and its learning objectives
- {{course_context}} — the course level and any principles or standards it must reflect
- {{evaluation_method}} — how correctness will be checked (e.g. unit tests, manual review, rubric scoring)
Instructions
- Ask for any missing inputs above before starting.
- Review {{assignment_description}} and identify the key learning objectives it should demonstrate.
- Build a grading rubric with clear, specific criteria for each part of the assignment, weighted appropriately for {{course_context}}.
- If {{evaluation_method}} involves automated checks, recommend specific unit tests or evaluation steps and how to set them up.
- Include guidance for grading partial credit and common edge cases.
Output format — A rubric table (criterion, description, points) followed by a short section on recommended tests or checks and how to run them. Clear and usable directly by an instructor or TA.
Guardrails — Do not assume grading standards not stated in {{course_context}}; ask rather than guess at institutional policy. Keep criteria specific and observable, not vague ("good effort"). Flag any part of {{assignment_description}} that is ambiguous and needs clarification before grading.
Example — {{assignment_description}}: a Python function that parses CSV files, with correctness and error-handling requirements; {{course_context}}: introductory programming course; {{evaluation_method}}: automated unit tests plus manual code-quality review.