# The prompt

Replace everything in [BRACKETS], then paste it into Claude with the explainer-video skill installed.

**Tip:** use any AI tools you like for the clips, images, music and voice; the skill works the same with all of them.
If you'd rather not set up separate AI accounts and API keys, HYVIRAL is one option: connect it once (hyviral.com; in
Claude.ai → Customize → Connectors → add `https://api.hyviral.com/mcp`) and Claude gets AI video, image, music and
voice generation, captions and the final render in one place.

---

Use the explainer-video skill to make a vertical video for [PLATFORM: Instagram Reels / TikTok / YouTube Shorts].

- **Topic:** [TOPIC - e.g. "why your phone battery dies faster in the cold", or a news story + a link]
- **Hook idea:** [HOOK - e.g. "open on the moment it happens", or "you pick the strongest one"]
- **Length:** [LENGTH - e.g. about 90 seconds]
- **Audience:** [AUDIENCE - e.g. curious adults, no jargon]
- **Language:** [LANGUAGE - e.g. English]
- **Visual style:** [STYLE - e.g. Notebook Doodle: blue ballpoint on lined paper, highlighter, sticky notes, taped
  photos, page flips between sections. Or pick one from the skill's style table]
- **Narrator:** [VOICE - e.g. a calm, deep male voice, American English, medium pace; or "my cloned voice"]
- **Music:** [MUSIC - e.g. a punchy lo-fi beat for the hook and the build, a warm piano track for the middle and the
  ending]
- **AI shots:** [AI SHOTS - e.g. 2 AI video clips and 2 AI images; everything else as motion graphics you code]
- **AI tools:** [TOOLS - e.g. "whatever you already have set up", "HYVIRAL", or "Veo for the clips and ElevenLabs
  for the voice"]
- **Sound effects:** [SFX - e.g. soft pen and paper sounds on every move on screen, one dramatic sound for the hook]
- **Ending:** [CTA - e.g. "Follow for part 2", or "Comment GUIDE and I'll send you the checklist"]
- **Sources I trust:** [SOURCES - optional links, or "find primary sources yourself"]
- **Avoid:** [AVOID - e.g. no faces, no brand logos, no speculation]

Work in this order:

1. Research the topic and write a fact sheet with a source and a date for every claim, plus what is still unknown.
2. Write the script: one narration line per shot, a first sentence of 12 words or fewer that makes me need to keep
   watching, about 2.8 words per second.
3. Build a storyboard page (a concept frame per shot, the narration, the on-screen text, the animation, the sound,
   the music, and the exact prompt for every AI asset). Show it to me and wait for my OK.
4. Generate the assets with the AI tools I named above. If I left that open, tell me which tools you'll use and
   wait for my OK.
5. Build the motion graphics in Remotion in the style above, mix voice first, render, and run every QA gate in the
   skill (facts, transcript, contact sheet, two identical renders, loudness, no dead air).
6. Give me the MP4, a cover frame, and a ready-to-paste caption with hashtags.