Learn

Give assignment feedback without writing for the learner

Turn rubric evidence into specific coaching while keeping revision decisions and prose with the learner

8 min setupTested with:ChatGPTReviewed: 2026-08-28
1

Add your context

Your text stays in this browser. AILesson Prompts does not send it to a model or server.

2

Your prompt

Unfilled fields remain visible as placeholders, so you can still copy and edit the prompt

Give specific assignment feedback as an educator without becoming the author.

Prompt, rubric, level, permitted support, and taught material:
[assignment]

Exact learner work, citations, prior feedback, stage, and questions:
[work]

Depth, priorities, tone, deadline, and authorship limits:
[policy]

Restate the assessed task and disclose missing information. For every judgment, cite a precise part of the learner's work and the applicable criterion. Separate demonstrated strengths, partial evidence, missing evidence, and matters that cannot be judged. Prioritize the few changes most likely to improve learning and rubric performance. For each, explain why it matters, ask a question that makes the learner reason, and name a revision operation; do not write replacement sentences, answers, evidence, personal reflection, or a submission-ready draft. Distinguish reasoning, content, evidence, structure, citation, and surface conventions. Avoid inferring effort, motivation, intent, or ability. End with a learner-owned revision sequence and a follow-up check that requires new evidence.
Private by defaultPrompt assembly happens locally in your browser. Avoid placing confidential information into any AI service unless your organization allows it.

From input to outcome

A worked example

See how concrete context turns this recipe into a usable result

Actual input

Assignment and rubric
Grade 9 science conclusion. Use the supplied plant-growth table to state a claim, cite two values, explain the pattern, and name one limitation. AI may give feedback but not compose text.
Learner work and context
More light makes plants taller. Group A was 12 cm and Group C was 19 cm. This proves light is always better. The experiment was accurate. Draft one; learner asks whether the evidence is enough.
Feedback policy
Warm, direct, under 350 words. Prioritize evidence reasoning and limitation. No replacement sentences or suggested final claim. Due tomorrow.

Example output

You have already cited two concrete values, which meets the quantity part of the evidence criterion. I cannot verify their accuracy because the table itself was not included.

Priority 1 — scope of the claim. Evidence: “This proves light is always better” extends beyond the reported A and C groups. The rubric asks for a pattern in the supplied table, not a universal rule. Question: what range of light conditions and plant type did this experiment actually test? Revision operation: limit the claim to the population, conditions, and measured outcome supported by the table, then explain how both values bear on that narrower claim.

Priority 2 — evidence reasoning. The two values are listed, but the draft does not explain why their difference supports the claim or whether the intermediate group follows the same pattern. Question: what happens across all groups, not only the endpoints? Revision operation: inspect every row, describe the direction and any exception, and connect that pattern to your own claim.

Priority 3 — limitation. “The experiment was accurate” is neither a limitation nor evidenced here. Question: which design feature restricts confidence—sample size, repeated trials, uncontrolled water, measurement precision, or something actually shown in the method? Revision operation: choose only a documented feature and explain what uncertainty it creates.

Revision order: verify the cited values against the table; define the tested scope; explain the full pattern; add one documented limitation; then check every rubric item. Bring the revised claim and its two-value explanation back for a second check; that new reasoning, not this feedback, should supply the final wording.

Why this works

  1. 1

    Evidence-linked comments are fairer and easier for a learner to act on.

  2. 2

    Questions and revision operations preserve authorship while teaching the process.

Check the result

  • Is every judgment tied to exact work and a rubric criterion?

  • Does feedback avoid ghostwriting and unsupported personal inference?

  • Does the follow-up require the learner to produce new evidence?

More ways to explore

Where this recipe fits

Text promptsFor learners & educatorsChatGPT prompts

Keep the work moving

Learn 9 min setup

Design a rubric from standards

Translate standards into observable, distinct, and consistently scorable criteria

A weighted analytic rubric with evidence rules, calibration cases, and a quality audit
Open recipe