Turn an idea into a spoken video script

Develop two honest opening hooks and a complete spoken script from an idea and factual material, with on-screen cues and timing estimates. A compact direction to help the next screen feel deliberate, specific, and easy to remix.

Write a complete spoken video script from an idea and factual material. Answer the question promised by the opening.

INPUT
Topic and main conclusion: [idea; state if the conclusion is still open]
Audience/use: [who watches, prior knowledge, long/short/tutorial format, desired understanding or action]
Material: [facts, cases, demonstrations, sources, genuine experience; separate unverified items]
Duration/language: [minutes, Chinese/English, speaking rhythm or measured pace]
Optional voice: [short personal sample, preferred expressions, tones to avoid]
Optional closing action: [one request; otherwise finish with the answer]

METHOD
1. Confirm topic, evidence and audience outcome. Ask up to three blocking questions and wait for missing core evidence or contradictions. Without a voice sample, use clear natural speech and never claim to have analyzed the creator's videos.
2. Draft five opening approaches, such as a specific question, demonstrable contrast, known result or real situation. Check whether the promise is deliverable, details are sourced and the topic is immediately clear. Select two and explain their different fit. Provide no fabricated tool scores or view predictions.
3. Write the full script for one opening. If no choice was made, name your recommendation and develop it. State the question and deliver useful information early. Give each beat one advancing point, with actual speakable sentences supported by examples, material or demonstrations rather than only an outline.
4. Insert separate [ON SCREEN: …] or [DEMO: …] cues where they aid understanding. Suggest visuals without generating media or a shot table. Return to the opening promise in the payoff. Use at most one main closing request.
5. Estimate and check timing. Prefer measured user pace. Otherwise use 220–280 Chinese characters or 130–160 English words per minute as rough drafting assumptions, explicitly requiring read-aloud calibration. Exclude screen cues from spoken counts and budget demos/pauses separately. If over time, suggest specific cuts rather than faster speech.

DELIVER IN ORDER
A. Two openings and recommendation: promise / supporting material / appropriate context; no tool scores.
B. Full script with estimated section times, spoken paragraphs and separate screen cues, connected naturally.
C. Production checks: actual spoken character/word count, pace assumption, demo/pause budget, total estimated range, unverified facts and the location of each promised answer.

BOUNDARIES
Invent no first-person experience, experiment data or audience feedback. Do not hide a missing answer behind suspense or guarantee reach. This is idea-to-script writing; use a storyboard template to adapt an existing script into shots.

Independent site adaptation of Jake Schincariol's yt-script (MIT). Uses opening, beat and payoff methods without its scoring scripts or personal voice files: https://github.com/Jakeschincariol/youtube-agent-skill/blob/main/skills/yt-script/SKILL.md .

invented experience, unsupported hooks, missing payoff, outline without spoken prose, fake hook scores, guaranteed reach, timing that ignores demos, automatic storyboard generation