faceless-videolisted
Install: claude install-skill popnow-ai/popnow-skills
# Faceless video workflow
A faceless video tells a story with narration over visuals, without anyone on camera. This workflow mirrors PopNow's Faceless Video Series app: it writes a script built for retention, voices it, generates one visual per beat, and hands back an ordered, captioned cut.
## What you deliver
- A script with a 1 to 2 second hook, numbered beats and an on-screen caption per beat.
- One voiceover track (or one per beat).
- One image or clip per beat in the requested aspect ratio.
- A caption file (SRT) built from the script and beat timings.
- Optional: one assembled MP4.
## Steps
1. **Brief.** Confirm topic, audience, platform, length (30 to 60 seconds for shorts, 3 to 8 minutes for YouTube), language, tone and whether visuals should be stills with motion or AI video clips. Video clips cost much more than stills; say so.
2. **Script.** Write it yourself: hook, setup, 3 to 7 beats, payoff, call to action. Keep each beat to one idea and about 6 to 12 seconds of speech. Only state facts you can support; mark anything uncertain for the user.
3. **Shot list.** For each beat write one visual prompt in a consistent style line (same palette, lens, era, art style) so the video feels like one piece.
4. **Quote everything** (see budget approval): the voiceover, every image, and every clip if using video.
5. **Voice.** Generate the narration with a TTS model; pick a voice that fits the niche and keep it fixed across a series.
6. **Visuals.** Generate stills (fast,