⚠️বাংলা (Bengali) content is governed and review-required. English fallback is shown where translations are pending reviewer approval.
Teacher Prompt Library
Reviewed, classroom-ready AI prompt templates
Every template uses the SATHI structure with marked variables, a filled example, its review requirements and its privacy note. Teacher review before classroom use is always required.
Review: Verify against real sources, never the AI itself; look up every citation; the approve/revise/reject verdict is recorded and yours. This is a SECOND pass: your own solving pass still happens; disagreements resolve against the textbook, never the AI.
Privacy: Never include student names, identifiable marks, health, family or community details. Describe situations and needs, never children.
Known failure modes: Both the key and the check can share the same error — the textbook is the referee
Verification & qualityPractitioner ~10 min Premium
Bias and inclusion audit of teaching content
Who appears, who is missing, who always leads — audited before students see it.
The full template and filled example are available with a premium plan.
Review: Verify against real sources, never the AI itself; look up every citation; the approve/revise/reject verdict is recorded and yours. Your students' own context appears first in the corrected version — check the flags.
Privacy: Never include student names, identifiable marks, health, family or community details. Describe situations and needs, never children.
Known failure modes: Over-correction into tokenism — variety should read naturally, not as a checklist
Verification & qualityPractitioner ~5 minFree
Extract every citation for manual lookup
A lookup-ready list of all references in a document — fakes look perfect.
Template
Extract every citation, reference, statistic and named source from this content into a lookup table: source as stated / the claim it supports / where I could check it (textbook, official site, library). Do NOT tell me whether they are real — formatting cannot prove existence; I will look each one up. Flag vague attributions ('studies show', 'experts say') separately as unverifiable-as-stated. Content: {{PASTE_CONTENT}}
See the filled example
Extract every citation and statistic from this AI-generated article into a lookup table with checking routes; flag vague attributions separately. [paste content]
Review: Verify against real sources, never the AI itself; look up every citation; the approve/revise/reject verdict is recorded and yours. A fake reference is worse than none: look up or leave out.
Privacy: Never include student names, identifiable marks, health, family or community details. Describe situations and needs, never children.
Known failure modes: The model asserts citations are real — the do-not-tell-me instruction keeps the judgement offline
Verification & qualityPractitioner ~10 minFree
Run the eight-step verification tree on content
Structured claim/source/age/bias/objective review of any AI output.
Template
Run a verification review of this content for Class {{CLASS}} {{SUBJECT}} (objective: {{OBJECTIVE}}): (1) list every FACTUAL CLAIM it makes, (2) for each claim, state how I could verify it against the textbook or an official source (do NOT verify by your own confidence), (3) list every citation/source it names — I will look each up, (4) age-appropriateness concerns, (5) bias/exclusion concerns (who appears, who is missing), (6) does it meet the objective? End with your uncertainty list: the claims you are least sure of. Content: {{PASTE_CONTENT}}
See the filled example
Run a verification review of this reading passage for Class 8 History (objective: causes of the revolt of 1857): claims listed, verification routes, citations listed for my lookup, age/bias notes, objective fit, uncertainty list. [paste passage]
Illustrative: twelve claims listed with verification routes; two citations flagged for lookup; one bias note (all urban examples); three claims on the uncertainty list.
Review: Verify against real sources, never the AI itself; look up every citation; the approve/revise/reject verdict is recorded and yours. The tree assists; the lookups and the verdict are yours.
Privacy: Never include student names, identifiable marks, health, family or community details. Describe situations and needs, never children.
Known failure modes: The AI 'verifies' its own claims confidently — the do-not-verify-by-confidence instruction blocks this framing, but the lookups remain yours
Verification & qualityPractitioner ~10 min Premium
SAFE review record for one output
The documented review that makes approval a decision — capstone evidence format.
The full template and filled example are available with a premium plan.
Review: Verify against real sources, never the AI itself; look up every citation; the approve/revise/reject verdict is recorded and yours. The blank judgement lines are the design: the record documents YOUR decision.
Privacy: Never include student names, identifiable marks, health, family or community details. Describe situations and needs, never children.
Known failure modes: The AI fills in the verdicts — the blank-lines instruction keeps judgement human
Verification & qualityPractitioner ~5 minFree
Score a prompt on the 10-point SATHI scorecard
Diagnose a weak prompt before regenerating — every flaw maps to a slot.
Template
Score this prompt on the SATHI scorecard — two points per slot (present? specific?), 10 total: S situation/subject, A audience/ability, T task/format, H human checks/boundaries, I iterate/improve. For each slot scoring under 2: name the missing element and write the one-line addition that fixes it. Then diagnose the overall disease if any: ambiguity / missing context / overconstraint. Prompt: {{PASTE_PROMPT}}
See the filled example
Score this prompt on the 10-point SATHI scorecard with per-slot fixes and the disease diagnosis: 'Make an engaging history activity for my class.'
Review: Verify against real sources, never the AI itself; look up every citation; the approve/revise/reject verdict is recorded and yours. The prescribed textbook wins for exam alignment even where richer sources exist.
Privacy: Never include student names, identifiable marks, health, family or community details. Describe situations and needs, never children.
Known failure modes: Web sources ranked above the prescribed text — exam alignment is the tiebreaker
Verification & qualityPractitioner ~10 min Premium
Three-version output comparison
Deliberate iteration: compare outputs from prompt versions and attribute differences.
The full template and filled example are available with a premium plan.
Review: Verify against real sources, never the AI itself; look up every citation; the approve/revise/reject verdict is recorded and yours. Save the version you approve — approval belongs to a version, not a prompt.
Privacy: Never include student names, identifiable marks, health, family or community details. Describe situations and needs, never children.
Known failure modes: Attribution guesswork if you changed multiple things per version — one change per version keeps it clean
Verification & qualityPractitioner ~5 min Premium
Weekly verification drill (planted-error passage)
Keep your verification eye sharp with a self-test — errors disclosed after.
The full template and filled example are available with a premium plan.
Review: Verify against real sources, never the AI itself; look up every citation; the approve/revise/reject verdict is recorded and yours. A teaching drill for you — never present the passage to students as real content.
Privacy: Never include student names, identifiable marks, health, family or community details. Describe situations and needs, never children.
Known failure modes: Planted errors too obvious — ask for 'subtle' if the drill stops challenging you
All example outputs are illustrative teaching fictions. AI drafts; the teacher remains responsible for accuracy, fairness, privacy and classroom fit.