Stable Diffusion for Fashion Designers: Build a Lookbook Prompt Kit

You are an expert AI fashion visual production specialist with 12 years of experience building Stable Diffusion prompt systems for fashion labels and independent designers where AI-generated lookbook imagery must meet the visual quality and stylistic consistency standards of professionally photographed editorial content without requiring a full photo production budget. Help me build a brand visual prompt kit so I can improve composition quality and produce a complete seasonal lookbook image set where garment detail, model presentation, and environmental styling are consistent across all shots and require minimal post-production correction. My situation: - Collection type and lookbook purpose: [e.g., "a 12-piece womenswear collection for an independent contemporary label — the lookbook will be used for wholesale buyer presentations, editorial submissions, and the brand's own e-commerce and Instagram channels"] - Target visual aesthetic: [e.g., "a quiet luxury editorial style — natural light, muted earth-tone environments, model poses that are relaxed and uncontrived, garment detail visible without styling overstatement — the reference aesthetic is Toteme and The Row's editorial photography approach"] - Garment description challenge: [e.g., "the collection is built around oversized tailoring, raw-hem detail, and natural fiber textures — linen, heavy cotton, and undyed wool — and previous Stable Diffusion attempts have produced synthetic-looking fabric texture and lost the raw-hem and natural fiber details in the image output"] - Model presentation requirements: [e.g., "the model should read as a young professional woman in her late 20s to mid-30s, with natural styling — no heavy editorial makeup, no dramatic pose — the garment is the visual subject, not the model"] - Environment types needed: [e.g., "three environments across the 12-shot lookbook — a minimalist interior (raw concrete or plaster wall, natural light from a large window), an urban exterior (clean brutalist architecture, overcast daylight), and a natural landscape (sparse grassland or scrubland, golden hour light)"] - Stable Diffusion setup: [e.g., "SD XL 1.0 with the Juggernaut XL checkpoint — currently running at 1024x1360 for portrait shots, 50 steps, CFG scale 7 — no ControlNet setup yet but open to adding it if it solves the garment detail problem"] - Output use and quality bar: [e.g., "images will be presented to wholesale buyers at A4 print size and used on the brand's website at 2000px height — the quality bar is editorial magazine standard, not social media snapshot standard"] Deliver: 1. A garment detail prompt block — a set of Stable Diffusion positive prompt terms specifically engineered for natural fiber fabric texture rendering, covering linen weave description, raw-hem edge language, oversized silhouette drape behavior, and the lighting description that reveals fabric texture without specular reflection — designed to be prepended to every lookbook shot prompt 2. Three environment master prompts — one for the minimalist interior, one for the urban exterior, and one for the natural landscape — each establishing the light source direction and quality, the background color temperature, the environmental surface texture, and the spatial depth that places the garment subject in the correct visual relationship to the background 3. A model presentation prompt block — a set of positive and negative prompt terms that together produce a natural, uncontrived model pose with accurate body proportions, no editorial makeup overstatement, and a gaze direction that keeps the viewer's attention on the garment rather than the face — designed to be appended to every shot prompt 4. Four shot-type prompt variants applicable to all three environments — a full-length standing shot, a three-quarter crop walking shot, a seated detail shot (emphasizing garment texture at close range), and a back-view shot (emphasizing garment construction and silhouette from behind) — each variant specifying the crop point, the pose language, and the garment detail focus 5. A ControlNet setup brief for garment pose consistency — a recommended ControlNet configuration using OpenPose for pose control and Tile for texture detail preservation, covering the model selection, the conditioning scale settings for each ControlNet unit, and the workflow for using a reference pose image to lock the model's body position across multiple shots in the same environment 6. A negative prompt master block for the full lookbook — excluding synthetic fabric texture, plastic or vinyl material rendering, dramatic makeup, harsh shadow, motion blur, background clutter, non-editorial color saturation, and any visual element that conflicts with the quiet luxury aesthetic 7. A lookbook shot sequence guide — a recommended generation order for all 12 shots covering which environment to establish first (interior, as it has the most controlled light and is easiest to achieve consistency within), how to carry the seed and ControlNet reference pose from one shot to the next within each environment, and how to review the full 12-image set for cohesion before submitting to the wholesale buyer presentation **Write every prompt block and variant assuming the garment is the primary subject of every image — all model, environment, and lighting decisions must serve the garment's visibility, texture legibility, and silhouette clarity, and any prompt element that draws attention away from the garment is a failure regardless of how aesthetically strong it is in isolation.**