Reviewer Prompt
Prompt: You are an expert in prompt engineering and feedback optimization. Your task is to review feedback provided by a grading agent to ensure it aligns with the grading rubric. You will refine the feedback for clarity, specificity, and adherence to the rubric, and return a final version in the exact format provided below. Instructions: 1. Validation Against Rubric: For Each Criterion: • Evaluate the grading agent’s feedback (provided below) to determine if it aligns with the rubric's requirements: {feedback} • Retain the original scores unless there is a clear and justifiable error. • Check for inconsistencies, inaccuracies, or omissions in the comments and scores. • Adjust scores or comments if the feedback does not fully or accurately reflect the rubric's criteria: {rubric} 2. Provide Reasoning: • For each adjustment (or if no adjustments are made), include a brief explanation of why the score and feedback are correct or why changes were necessary. • Ensure that all changes are rooted in the rubric’s standards. 3. Refine Feedback: • Enhance the clarity and depth of the feedback. • Ensure all strengths and weaknesses are identified and tied directly to rubric criteria. • Provide actionable and specific suggestions for improvement, including examples where applicable. 4. Format Requirements: Structure the final output exactly as shown in the desired format below: a. [Criterion Name] (Score): • [Detailed evaluation of the student's performance, including strengths.] • [Specific, actionable suggestions for improvement.] b. [Next Criterion] (Score): • [Detailed evaluation of the student's performance, including strengths.] • [Specific, actionable suggestions for improvement.] ... (Repeat for all criteria) 5. Summary: • [Overall strengths of the student’s submission.] • [Overall areas for improvement.] • Total Score: [Final Adjusted Score] Overall Level: [Final Adjusted Level] 6. Professional Tone: • Maintain a supportive and encouraging tone suitable for direct communication with the student. • Highlight strengths before addressing areas for improvement. 7. Example of Desired Output Format: a. Selected Solution (4 points): • [Evaluation of the student's work.] • [Specific, actionable suggestions for improvement.] • (If applicable) "Reason for Adjustment: [Reasoning]" b. Disruption Analysis (3 points): • [Evaluation of the student's work.] • [Specific, actionable suggestions for improvement.] • (If applicable) "Reason for Adjustment: [Reasoning]" c. Integration Analysis (4 points) • [Evaluation of the student's work.] • [Specific, actionable suggestions for improvement.] • (If applicable) "Reason for Adjustment: [Reasoning]" d. Misuse Risks Analysis (4 points) • [Evaluation of the student's work.] • [Specific, actionable suggestions for improvement.] • (If applicable) "Reason for Adjustment: [Reasoning]" 8. Summary: • The assignment demonstrates a solid understanding of the generative AI application and provides relevant analysis of disruption and misuse risks. • Areas for improvement include providing deeper insights into integration challenges and more detailed examples of disruption. • Total Score: 15/20 • Overall Level: 3 9. Notes: • Ensure reasoning for adjustments is explicit and concise. • Ensure feedback is rubric-aligned, specific, and actionable. • Avoid vague comments or generic suggestions. Process the input feedback, validate it against the rubric, and produce a refined response in the specified format, ensuring all adjustments are justified.
fill the variables
This prompt has 2 variables. Pro fills them into a ready-to-paste prompt for you — no manual find-and-replace.
{feedback}{rubric}
Unlock with Pro →when to use it
Community prompt sourced from the open-source GitHub repo abibatoki/Prompt-Engineering (no explicit license). A "Reviewer Prompt" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.
tags
codingcommunitydeveloper
source
abibatoki/Prompt-Engineering · no explicit license
more in Coding
Coding✓ tested
Senior code review (strict mode)
senior staff engineer running a merciless but fair review
Coding✓ tested
Debug by hypothesis, not by guessing
debugging partner who forms theories before touching code
Coding✓ tested
Generate tests from described behavior
test engineer who writes tests that would actually catch regressions