Create a high-fidelity desktop application UI for “Salvor AI”, an AI identity-to...

Create a high-fidelity desktop application UI for “Salvor AI”, an AI identity-to-intent creation studio. Use the supplied original Standard UI as the primary structural reference. The core concept is: IDENTITY → INTENT → CREATE Users upload photographs to create reusable identities of themselves, friends, family, pets, characters, or other subjects. They then visually construct an intent for those identities: a different place, era, outfit, hairstyle, appearance, action, pose, environment, style, image, or video. There is NO traditional prompt-writing experience. The interface itself is the creative language. IMPORTANT: The UI must be simple enough for a first-time user, but it should NOT feel boring, sterile, or like a generic settings form. Make it feel like a small creative playground. BALANCE: Use a mixture of interface controls and carefully selected imagery. Do not create a gallery-heavy interface and do not fill the screen with photographs. Use imagery for: Identity thumbnails A limited number of intent presets Small visual references The generated result Subtle contextual previews Use controls for most interaction.

VISUAL Dark cinematic navy/charcoal application, sophisticated violet/purple accent, subtle blue highlights, soft borders, rounded panels, modern typography, gentle depth, polished micro-interactions. Premium and imaginative, but calm. No pricing, upgrade buttons, subscriptions, advertisements, or sales messaging. LAYOUT: HEADER: Salvor AI branding, current creation name, understated utility controls, and a clearly visible Advanced toggle. Advanced is OFF. LEFT — IDENTITIES: A compact “Identities” area containing a few small identity thumbnails and: Create Identity The primary empty state should be inviting: “Who are you creating?” “Upload photos to create a reusable identity.” Include Upload Photos / Take Photo. Once created, identities become reusable visual building blocks. MAIN CREATION AREA: Start with: “Bring an identity to life.” Show the selected identity as a compact visual card. Then present: “What do you want to create?” Instead of a large form, provide a small selection of visually interesting starting points. PRESET INTENTS: Approximately 6–8 small cards: Renaissance Portrait Tokyo After Dark Summer Escape Medieval Adventure Old Hollywood Future Explorer Editorial Portrait Surprise Me Each has a small atmospheric image and concise title. They are starting points, not giant gallery tiles. Selecting one instantly populates the intent controls. Below this, show a clean horizontal or stacked set of creative controls: WHERE Florence · Tokyo · Beach · Mountains · Custom WHEN Modern · 1920s · Renaissance · Medieval · Future LOOK Current · Formal · Streetwear · Historical · Custom APPEARANCE Hair · Tattoos · Accessories · Other DOING Portrait · Walking · Sitting · Adventure · Interaction STYLE Photorealistic · Cinematic · Vintage · Fantasy OUTPUT Image · Video Make the currently selected choices visually prominent. CREATIVE FEEDBACK: As selections change, show a small “Intent” summary that dynamically updates. For example: ME + RENAISSANCE FLORENCE + ROYAL CLOTHING + PORTRAIT + CINEMATIC This should feel like the creation is taking shape. Where appropriate, show a tiny visual transition or preview thumbnail reflecting the selected choices. Include a subtle “Surprise Me” action that can generate an unexpected but coherent combination of intent attributes. CONTEXTUAL CONTROLS: Every major category has a subtle sliders/settings icon. Opening it produces a compact flyout with deeper options. Hair: Style · Length · Color Appearance: Tattoos · Makeup · Facial Hair Clothing: Material · Color · Accessories Pose: Body Position · Gaze · Expression Location: Weather · Time · Environment

Camera: Framing · Lens · Depth of Field Identity: Preservation · Feature Locks These flyouts should feel like discovering hidden creative possibilities. RIGHT — RESULT: Keep a meaningful output area without making it dominate the entire application. Before generation, show an attractive “intent preview” using the selected identity and a restrained visual representation of the chosen world/style. The preview should communicate anticipation rather than simply being a blank canvas. After generation, show the resulting image or video. Primary action: GENERATE Nearby: Variations · Remix · Change Intent · Create Video · Save SOCIAL: A compact option for: Instagram · TikTok · YouTube · Pinterest · Profile Do not turn social into a major section. ENGAGEMENT: The application should subtly feel alive. Use tasteful micro-interactions: Selected controls glow softly in violet Intent summary updates immediately Preset selection changes contextual imagery Small transitions between identity and intent Generate button becomes more visually confident as the intent becomes complete Selected attributes appear as elegant chips Small “Surprise Me” moments introduce discovery Empty states feel like invitations rather than forms Do NOT use excessive animation, flashy gradients, gamification, badges, or distracting effects. ADVANCED: The header contains an Advanced toggle. Switching it ON preserves the exact same identity and intent and reveals deeper controls in place. The Standard interface should feel like the inviting surface of a much more capable creative system. The emotional goal is: “I came here with an idea.” → “I wonder what I could make.” → “Oh, that's interesting.” → “Let's create it.” It should feel creative and inspiring without becoming content-heavy. Desktop 16:9, production-ready SaaS application, high-fidelity UI, elegant dark creative software, balanced imagery and controls, sophisticated UX, cinematic but restrained, no browser chrome.