Turn verified material into a timed, visual, spoken script that teaches one useful idea
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
Write a production-ready informational short-video script using only the supplied material.
Facts, explanations, examples, locators, dates, permissions, and limits:
[material]
Audience, objective, platform, duration, aspect ratio, presenter, tone, CTA, and language:
[brief]
Footage, locations, graphics, audio, captions, brand, safety, budget, and approvals:
[production]
Choose one learning promise that fits the duration. Build a timed beat sheet with hook, context, explanation, concrete demonstration, limitation or boundary, recap, and proportionate CTA. For every beat specify seconds, spoken words, on-screen text, shot or graphic, source locator, audio cue, and production dependency. Write for speech, not an essay; ensure total words fit the speaker rate and leave visual breathing room. The hook must be accurate without withholding a crucial qualifier. Do not invent footage, demonstrations, statistics, outcomes, people, quotations, product behavior, or trends. Label fictional demonstrations. Make the message understandable with captions and make essential meaning independent of color or sound. End with word-count timing, asset gaps, claim audit, and final production checklist.
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
Source material
Approved article: visibility means seeing work state; ownership means authority to decide readiness and acceptance. A status tool can display roles but does not itself assign accountability. Four audit questions: who declares ready, who accepts, what evidence counts, who resolves dispute. Fictional example allowed if labeled. No performance statistics.
Video brief
60-second vertical LinkedIn video for small-team operations leads. One presenter, calm and practical, about 130 spoken words/minute. Objective: viewers can name the four audit questions. CTA: save and audit one handoff. English and Chinese example output.
Production constraints
Office desk, phone camera, presenter, simple text cards and arrows. No customer footage, stock video, voice clone, or music license. Burned-in captions required. Brand colors blue and white, but meaning must not rely on blue. One afternoon shoot; no legal review if claims remain in article.
Example output
0-6s — Presenter holds a card marked READY beside a second card marked “Who accepts?” Spoken: “A green status can show where work is. It cannot tell you who has authority to accept it.” On-screen: “Visibility ≠ ownership.” Source: definitions. No music dependency.
6-14s — Two labeled boxes and an arrow. “Before adding another field, audit one real handoff with four questions.” On-screen numbered 1-4.
14-24s — Move READY card. “One: who declares the work ready? Two: who accepts it on the other side?” Source: audit questions 1-2.
24-36s — Show a checklist card. “Three: what evidence proves acceptance—not just activity? Four: who resolves a disagreement?” Source: questions 3-4.
36-47s — Clearly label “Fictional example.” “A report can sit at ready because the sender expects upload and the reviewer expects approval. The missing part is a decision rule, not another color.” Source: permitted fictional demonstration plus definitions.
47-56s — Presenter points to all four cards. “Name the declarer, accepter, evidence, and resolver. Then choose how a tool should display them.” Source: synthesis; retains the boundary that tools do not assign accountability.
56-60s — “Save this and audit one handoff this week.” On-screen CTA and four-word recap.
Timing: about 125 spoken words, appropriate for roughly 58 seconds at 130 words per minute with short pauses. All essential labels are captioned and numbered, not color-coded alone. Assets: five printed cards, phone tripod, caption file. No unsupported result or real customer example appears.
Why this works
1
A timed audiovisual contract exposes scripts that cannot fit or be produced.
2
Beat-level sourcing keeps fast hooks from outrunning the evidence.
Check the result
Do words, pauses, and shots fit the stated duration?
Can every factual beat be traced to a supplied source?
Can viewers understand essential meaning with captions and without audio?
ChatGPT