screening and conversion.md
---------

SCREEN RECORDING → 9:16 SHORT-FORM VIDEO — INSTRUCTIONS FOR [APP NAME]

PHASE 1 — RECORD (clean, no effects)
- Capture native app screen at the highest resolution available; minimum 1080x1920, prefer ~1.3x the delivery target (e.g. 1284x2778+) so later zooms have room to move without upscaling.
- No filters, overlays, color grading, or effects during capture.
- If the device records in HDR, tone-map to SDR before anything else in this pipeline touches the file.
- Confirm no critical UI (nav bar, key numbers, buttons) sits within 10% of any frame edge; note which on-screen element each screen should later zoom toward.
- Archive the raw file before processing. Never overwrite it.

PHASE 2 — PROCESS (automatic, in this order)

1. COLOR GRADE (fixed preset — do not re-derive per video)
   - Style: clean high-key / soft neutral. No vignette.
   - White balance: ~5600K
   - Exposure: +0.2 to +0.3 EV
   - Black level: lift blacks to RGB 8-10
   - Highlights: cap whites at RGB 245
   - Saturation: -5% to -8%
   - Grain: 0% default; up to 3% only if a screen feels sterile — never on numeric/result close-ups
   - [Replace with the app's actual locked brand values once defined]

2. MOTION (Ken Burns / punch-ins)
   - 1.10x-1.15x zoom on major screens only (results, calculations, key numbers)
   - Zoom target = the specific number/result, never the full screen
   - Ease in/out — no linear zoom
   - Zoom completes over the full duration the screen is on camera

3. SAFE ZONE (applies to every caption and graphic below — placement must respect this from the start, not be checked afterward)
   - Canvas: 1080x1920
   - Keep all text, logos, and key graphics inside: x: 60-930, y: 250-1450
   - Covers TikTok, Instagram Reels, and YouTube Shorts UI with margin
   - Confirm on an actual phone screen before publishing — not desktop preview

4. CAPTIONS
   - Burn in, synced to voiceover
   - Group in 2-4 word phrases, not single words
   - Minimum hold: 0.4-0.5s per chunk, even if speech is faster
   - First 3 seconds = hook: distinct, bolder treatment (single punchy line or number reveal), must read with sound OFF
   - Steady-state caption style resumes after the hook

5. NUMBERS & PATTERN INTERRUPTS
   - Calculated results (BMI, calories, scores, etc.): animate as count-up from 0 to final value, 0.5-0.8s, synced with pop SFX — never a static pop-in
   - Highlight circles / soft glow: 1-2 per screen max
   - Soft whoosh SFX on screen transitions only

6. AUDIO
   - Voiceover: primary, mixed forward
   - Music bed: on-brand/trending, ducked 8-12dB under voiceover during narration
   - SFX (whoosh/pop): at least 15dB below dialogue
   - Normalize to -14 LUFS integrated, -1 dBTP true peak
   - Optional: -11 to -12 LUFS specifically for TikTok/Reels exports if prioritizing feed attention over consistency with YouTube

7. EXPORT
   - 1080x1920, H.264 High Profile
   - 30fps default; 60fps only if source has fast scroll/swipe motion
   - 15-20 Mbps
   - MP4

PHASE 3 — QA (before publish)
[ ] Muted playback — hook and captions fully understandable with no audio
[ ] Real phone preview — nothing sits under platform UI
[ ] Grade matches locked preset — no drift from prior exports
[ ] No distorted or illegible numbers/text anywhere in the video