explainer-video-production

What it does

Produce an explainer only after defining what the audience should understand, believe, decide, or do differently after watching. Treat the video as a learning and decision artifact, not as a sequence of pretty scenes.

This skill is provider-independent. Use the available generation, editing, composition, captioning, and review tools in the host environment, but keep the explainer logic, factual discipline, accessibility, and QA standards intact across providers.

How it works

Follow non-negotiables: start from audience, objective, and measurable outcomes; separate documented facts, inferences, production heuristics, and creative inventions; maintain a claim log for various claim types; escalate high-risk claims to appropriate reviewers; build accessibility into script, storyboard, captions, audio, data visuals, and delivery variants; re-check volatile provider/platform facts at production time. Do not ask image/video models to invent evidence or rely on exact legal/medical statements. Use deterministic layout for text and charts.

Intake asks for audience, objective, use case, success evidence, constraints, risk tier, and a strong objective written as an observable outcome.

Research and citation discipline requires a claim matrix before scripting with fields for claim, type, source, evidence strength, on-screen handling, risk, and owner; prefer primary sources and label secondary when used. Include provenance for AI assets when applicable.

Decompose the concept into atoms (terms, parts, mechanism, contrast, example, consequence, action) and choose a minimal necessary set.

Script architecture for most 60-180 second explainers: Hook, Promise, Map, Build, Example/demo, Caveat, Action. Include variations for short social, training/onboarding, and public-service formats.

Write narration to aid comprehension: simple language, one idea per sentence, signposting, consistent terminology, contextualized numbers, and plan 130-160 words per minute as a heuristic. Use analogies only when mapping is accurate and safe.

Storyboard each scene with columns covering time, learning beat, narration, visuals, motion, captions, evidence, accessibility, and asset notes; ensure visuals support the learning goal and use a consistent color language for problem/solution/warning/proof.

For diagrams and data, choose simple charts, state takeaways in words, and ensure accessibility alternatives exist.

Motion should reveal change and support explanation without distracting from content, and narration should dominate over music or SFX.

Captions, transcripts, and accessibility are built in from the script stage; provide descriptive transcripts and ensure captions cover speech and meaningful non-speech audio.

When to use it

Use when translating a topic, brief, research corpus, product, process, policy, dataset, or complex idea into a factual, accessible, narrated, captioned, storyboarded, generated-media-ready explainer video with claim review, localization, pacing, variants, and QA.

What it can touch

The skill requires tools in the host environment for generation, editing, composition, captioning, and review; declared tools include claude-code, codex, copilot, cursor.

Caveats

Escalate high-risk claims to the client or SME and do not provide professional advice (legal, medical, financial). Maintain separation of facts and inferences; re-check volatile facts at production time; do not rely on image/video models to invent evidence or render precise technical statements. Ensure accessibility targets are met and document sources in the claim matrix.