01Ming × Versailles
⭐ Storyboard MCP · Fusion flagship
THE ROBE COAT one silhouette. twelve courts.
A single floor-length open-front robe-coat silhouette — same cut, same bell-sleeve drape, same model pose, same key-light direction held constant — rendered through twelve curated east-meets-west royal-court fusion lenses. Ming silk damask meets Versailles court ornament. Heian junihitoe meets Elizabethan ruff. Safavid filigree meets French Renaissance pearl. The fashion-week capsule presentation no single atelier could produce in a season.
The 12 looks — click to view full resolution
Every look uses the same robe-coat silhouette spec, the same 85mm full-figure runway shot, the same cool single-direction key light from upper-front at 20°. The SILHOUETTE is the through-line; the cultural textile + ornament + accessories are the variable. The viewer reads the gallery as one capsule garment in 12 ornament treatments, not 12 unrelated dresses — that's what makes it defensible to a fashion creative director.
02Heian × Elizabethan
03Safavid × French Renaissance
04Mughal × Spanish Baroque
05Edo × Belle Époque
06Joseon × Tudor
07Tibetan × Byzantine
08Mongol × Russian Imperial
09Khmer × Greek Doric
10Rajput × Pre-Raphaelite
11Tang × Napoleonic Empire
12Ottoman × Sargent
Why one-silhouette-twelve-courts is the load-bearing creative choice. Fashion development pitches benefit from seeing the SAME silhouette explored across cross-cultural ornament traditions — it shows the silhouette has structural strength while the ornament + textile vocabulary is being interrogated. By hand that requires deep textile + ornament training in every culture — weeks per look. This collapses it to one afternoon: pick the silhouette, pick the pairings, run.
The 60-second runway reel — each look turns slowly to face camera
Each look turns very slowly from three-quarter to full-front, eyes meeting camera at the end — drape stays still, no aggressive fabric or wind. Scored under by a low-pulse bass + bright string-flourish runway score, the rhythm of a Paris fashion-week capsule reveal.
The runway · 1080×192060.0s · 12 looks × 5s · runway score
Fusion grammar — what gets blended where
01
Ming × Versailles
Cinnabar silk damask + phoenix embroidery meets gilded brocade trim + pale gold satin lining + Louis XIV court-tier construction.
02
Heian × Elizabethan
12-layer junihitoe gradient palette meets subtle ruff at collar + farthingale-hip fullness + slashed-sleeve embroidery.
03
Safavid × French Renaissance
Cobalt + saffron Persian floral filigree meets Catherine de Medici pearl-encrusted collar + slashed contrasting sleeves.
04
Mughal × Spanish Baroque
Emerald silk + gold zardozi metalwork meets Velázquez wine-and-black velvet palette + brocade panels.
05
Edo × Belle Époque
Indigo Hokusai wave-pattern textile meets fitted Worth-house bodice + bustle silhouette + art-nouveau iris cuff embroidery.
06
Joseon × Tudor
Apricot hanbok jeogori + navy chima palette meets Henry VIII jeweled rectangular collar + gold-thread strapwork.
07
Tibetan × Byzantine
Crimson dharma-wheel brocade meets gold-tesserae imperial dalmatic bands + purple imperial-tier shoulders.
08
Mongol × Russian Imperial
Navy cloud-and-crane embroidery + felt-and-leather shoulders meets Catherine the Great gold-pearl epaulets + fur trim.
09
Khmer × Greek Doric
Stone-warm apsara silk + Khmer gold collar meets vertical pleated chiton drape + Greek-key meander hem + fibula brooch.
10
Rajput × Pre-Raphaelite
Maroon Pichwai peacock filigree + shisha mirror-work meets Burne-Jones chiffon drape + foliate gold crown band.
11
Tang × Napoleonic Empire
Turquoise Tang cloud-and-bird embroidery meets Empire-waist gold-laurel band + narrow column drape below.
12
Ottoman × Sargent
Teal Ottoman tulip-and-leaf bullion meets Sargent's Madame X commanding shoulder + silver-grey satin sash + black-velvet revere.
Pipeline trace — what the storyboard MCP actually did
Wave 1 — silhouette constant + 12 royal-court fusion looks
12 ×
Wave 2 — slow turns + runway score 12 ×
Wave 3 — local stitch ffmpeg normalize: kling 1108×828 → 1080×1920 portrait (scale + crop) ffmpeg concat: 12 clips → single 60s timeline (concat demuxer, no re-encode) ffmpeg mix: runway score @ 0.95 vol + clip ambient ducked to 0.22, afade in/out output:
Quality controls applied ✓ Garment silhouette constant repeated VERBATIM in every look (shape + drape stability) ✓ Composition + pose constant (85mm, runway pose, key light upper-front 20°) verbatim in every look ✓ Same model proportion across all 12 looks (reads as ONE garment, 12 ornament treatments) ✓ Curated "X textile × Y ornament + Z accessory" fusion grammar — no kitsch ✓ NO readable text guard (defense against AI text-rendering failures on labels/tags) ✓ Fabric drape plausibility constraint ✓ Slow turn only on i2v (no aggressive fabric or wind motion that would break editorial aesthetic) ✓ 25/25 jobs succeeded first try · ZERO moderation fallbacks · ZERO infra errors
create_media action=generate, model=gpt-image, aspect_ratio=9:16 → 12/12 done in ~115-130s each, ZERO moderation flags
Each prompt = same silhouette constant + same 85mm runway composition + fusion-specific block. Silhouette + pose stability bias works.
Wave 2 — slow turns + runway score 12 ×
create_media action=animate, model=kling-v3-i2v, duration=5, prompt="slow turn to face camera, drape still, eyes meet at end" → 12/12 done in ~49-71s each
1 × create_media action=music, model=minimax-music, prompt="low pulsing bass + bright string flourishes on each look transition" → done
Wave 3 — local stitch ffmpeg normalize: kling 1108×828 → 1080×1920 portrait (scale + crop) ffmpeg concat: 12 clips → single 60s timeline (concat demuxer, no re-encode) ffmpeg mix: runway score @ 0.95 vol + clip ambient ducked to 0.22, afade in/out output:
the-robe-coat-reel.mp4 (41MB, 60.0s)
Quality controls applied ✓ Garment silhouette constant repeated VERBATIM in every look (shape + drape stability) ✓ Composition + pose constant (85mm, runway pose, key light upper-front 20°) verbatim in every look ✓ Same model proportion across all 12 looks (reads as ONE garment, 12 ornament treatments) ✓ Curated "X textile × Y ornament + Z accessory" fusion grammar — no kitsch ✓ NO readable text guard (defense against AI text-rendering failures on labels/tags) ✓ Fabric drape plausibility constraint ✓ Slow turn only on i2v (no aggressive fabric or wind motion that would break editorial aesthetic) ✓ 25/25 jobs succeeded first try · ZERO moderation fallbacks · ZERO infra errors
Cost — total MCP run end-to-end
Wave 1 looks
~$2.40
12 × gpt-image @ ~$0.20
Wave 2 turns
~$6.84
12 × kling-v3-i2v @ ~$0.57
Music score
~$0.40
1 × minimax-music
Total
~$9.64
on-target for the $10 flagship
Reproduce this — the playbook is open
This reel was built from the fusion-garment-runway playbook — the same template can produce ANY single-silhouette capsule: a fitted high-collar dress across 12 court traditions, a wide-sleeved tunic across 12 textile vocabularies, a structured coat across 12 ornament languages. Two sister playbooks: fusion-character-portrait (one face — see THE EMISSARY) and fusion-architecture-walk (one building — see THE GATE).