Teachers · Instructional coaches · 15 suggested minutes
Turn formative evidence into a next step
Use a small evidence pattern to compare teaching options while keeping real records local and decisions with the teacher.
Before you begin
- Use school-approved tools and follow local policy. Keep student work, names, identifiers, photographs, voices, disability information, behavior records, and staff or personnel records out of prompts. Removing a name does not necessarily make a record anonymous.
- Use the supplied synthetic inputs when trying a prompt. If you consider a real aggregate later, have an authorized person check that it cannot identify anyone and that the intended tool and use are permitted. Small groups and unusual details can identify people. When unsure, keep the evidence offline and use a fictional pattern.
- Teachers decide what to teach and how to respond. Do not use these workflows for automated grading, placement, discipline, eligibility, teacher evaluation, or other consequential decisions.
- Do not record lessons or coaching conversations for these workflows. Keep planning notes brief, teacher-owned, and separate from personnel evaluation. Do not score teachers or monitor individual participation.
- Treat every AI answer, answer key, citation, and standards claim as an unchecked draft. Verify facts and calculations using sources you can inspect. Standards alignment requires the exact official standard and a teacher review; it is not claimed here.
- Use only materials you have permission to share with the tool. A public web page is not permission to upload or reproduce its entire contents.
Bring these inputs
- The original task, intended learning, and teacher-checked answer
- For an AI trial, the fictional aggregate in the example below; for real planning, inspect evidence locally under school policy
- The time and materials available for the next lesson
The output to keep
Two plausible interpretations, a teacher-selected short teaching response, and a fresh check on the same idea.
Try the workflow
15 suggested minutes in total. Allow more time whenever checking, access or local policy requires it.
Describe what is visible 3 suggested min
Inspect the task before interpreting responses. Privately sort the evidence by what a response shows, not by labels for learners. Check totals and missing responses. For the prompt, use a synthetic pattern or an explicitly permitted, genuinely non-identifying aggregate.
Separate observation from explanation 3 suggested min
Write what the evidence shows and at least two things it cannot establish. A wrong answer alone does not identify the cause; a correct answer without a method may leave the reasoning unclear.
Compare two next steps 4 suggested min
Use the prompt to generate two short options. Compare their fit with the goal and the evidence. Keep diagnosis, individual grouping, and consequential decisions out of the model’s role.
Choose and check 3 suggested min
Select the response yourself. Solve its examples and the fresh check. Decide what you will look for locally before changing the next lesson again.
Keep a small revision note 2 suggested min
Record the selected teaching move and one limitation of the evidence. Keep any actual student information in its approved school location, separate from this workflow log.
Copy-ready prompt
Replace the placeholders with public, teacher-authored or invented material. Keep personal information out of the prompt.
Help an educator consider a next lesson using this synthetic evidence only. Learning goal: [goal] Task and checked answer: [task and answer] Fictional aggregate: [invented response counts or broad fictional pattern] Available time/materials: [constraints] Separate observations from possible explanations. Give two plausible interpretations and what further evidence would distinguish them; do not diagnose students. Suggest two brief whole-class teaching options, explain the trade-off, and give one fresh check with a worked answer. State what cannot be concluded. Do not assign grades, identify learners, recommend placement or discipline, or infer ability, disability, motivation, or demographics. The teacher will select and verify the next step.
Invented example · not an outcome claim
Synthetic worked example
Context: All counts below are invented. They describe 24 fictional responses, not an actual class.
Input: Task: calculate 3/4 − 1/8 and show a method. Goal: subtract fractions with unlike denominators using equivalent fractions. Fictional categories: 11 responses give 5/8 with a valid method; 6 give 5/8 without a method; 7 give 2/4 without a clear method. The categories total 24.
Example output
- Observation: 11 responses visibly demonstrate the intended method. The 6 correct answers without methods do not show how they were obtained. The 7 other answers do not establish a shared misconception.
- Possible explanations to investigate: some learners may not be using equivalent fractions, or the directions may not have made the expected explanation clear. The counts cannot distinguish these possibilities.
- Option A: spend 5 minutes comparing two teacher-created methods, 3/4 − 1/8 = 6/8 − 1/8 = 5/8 and the incorrect 2/4. Ask which preserves the size of the fractional parts and why. Option B: model the calculation with same-whole eighths strips, then connect the drawing to symbols. A emphasizes reasoning about methods; B makes the units visible.
- Fresh check: 2/3 − 1/6. Teacher-checked answer: 4/6 − 1/6 = 3/6 = 1/2. Ask for a diagram or written reasoning, then inspect locally whether the equivalent-fraction step is visible.
The human decision: The fictional teacher chooses Option B, then uses the fresh check to decide whether another example is needed. This is a provisional teaching choice, not a finding about any learner.
Teacher-owned checks
- Is the task itself clear, accessible, and correctly keyed?
- Do the categories avoid double-counting, and have I accounted for missing evidence?
- Have I separated what a response shows from what I am guessing?
- Are all real records kept in approved locations, with no identifying detail sent to an AI tool?
- Does the next check ask about the same idea without simply repeating the answer?
Measure this attempt
Record minutes spent preparing the next step, whether you produced two distinct options, and whether a fresh check was verified. Do not turn response counts into a claim that AI caused a learning gain.
Repeat deliberately
After teaching, review the new evidence locally. Keep, revise, or drop the teaching move based on your judgment and the task’s limitations. Repeat with a new synthetic pattern if you want AI help; do not accumulate student profiles in chat.
Adapt the method
- In literacy, use fictional patterns such as “claim stated, evidence missing.” In science, use “observation supplied, explanation unclear.” Avoid interpreting an error as a diagnosis.
- Low-prep version: choose one common response pattern, draft one clearer example yourself, and check it with a colleague.
- Optional support: permit a diagram or spoken explanation when the learning goal does not require extended writing.
Your check-and-repeat log
Keep this optional log privately, on paper or in an approved teacher-owned file. Record process details only. Do not add names, student work, response-level data, coaching transcripts, recordings, or personnel judgments. A few lines are enough; no submission or score is required.
- Date and workflow used
- Space for your own offline notes
- Small planning task and intended output
- Space for your own offline notes
- Actual minutes spent, including checking
- Space for your own offline notes
- What I independently checked and where
- Space for your own offline notes
- One correction or unresolved issue
- Space for your own offline notes
- Decision: keep / revise / drop / hold
- Space for your own offline notes
- One change to try next time
- Space for your own offline notes
- Optional next reflection point
- Space for your own offline notes
These are blank spaces for your offline copy. This site does not collect or store entries.
Public sources & their limits
- Guidance for generative AI in education and researchUNESCO
Supports: UNESCO emphasizes human agency, data protection, and educational and ethical review of generative AI use. These principles inform the toolkit’s adult-led use and verification boundaries.
Limit: Policy guidance does not validate these prompts, certify a particular tool, or establish compliance with local rules.
- Teacher Feedback to Improve Pupil LearningEducation Endowment Foundation
Supports: The guidance focuses on feedback grounded in teaching and an understanding of learning gaps. This informs checking the learning goal and using evidence to consider a next teaching step.
Limit: The source concerns teacher feedback. It is not evidence that AI analysis, these prompt sequences, or the suggested timings improve learning.
Published October 13, 2026. Times are suggested planning estimates, not measured completion times. These workflows have not been classroom-tested, and no learning gains or time savings are claimed. Public primary-source guidance informs selected design choices; it does not establish this toolkit’s effectiveness. Process measures help an educator decide whether to keep or revise a workflow. They are not student-outcome measures or teacher-performance scores.