{"id":42780,"plugin_id":"plugin_asdk_app_6a8c52a06c3081919407e081a375c210","kind":"skill","collection_source":"plugin_package","comparison_source":null,"observed_at":"2026-10-24T13:07:05.923Z","digest":"fed83542d9224811a89e8ed57bf36c18db3c6396532d1eaa12eea8b8f5fbce3f","against":null,"payload":{"description":"Create a surreal cinematic VFX shot where a character walks and looks around as the environment rises and folds inward into a vast gravity-defying enclosure; use for folding cities, bending landscapes, enclosing architecture, world-fold transitions, ambiente che si piega, città che si solleva, paisaje plegable, or gravity-defying environmental transformation, not for subject deformation, generic camera rolls, portals, or ordinary scene transitions.","included_files":[],"name":"vfx-foldscape","skill_md_contents":"---\nname: vfx-foldscape\ndescription: Create a surreal cinematic VFX shot where a character walks and looks around as the environment rises and folds inward into a vast gravity-defying enclosure; use for folding cities, bending landscapes, enclosing architecture, world-fold transitions, ambiente che si piega, città che si solleva, paisaje plegable, or gravity-defying environmental transformation, not for subject deformation, generic camera rolls, portals, or ordinary scene transitions.\n---\n\n# Foldscape VFX\n\n## Tool requirements\n\n- **Core evidence and bindings:** `prostudio_brief_get`, `prostudio_workspace_get`, `prostudio_references_list`, `prostudio_references_resolve`, `get_prostudio_media_catalog`, and `inspect_prostudio_media`.\n- **Final Foldscape clip:** `prostudio_models_list`, `prostudio_generation_prepare`, `prostudio_video_generate`, `prostudio_job_status`, and `prostudio_generation_get`.\n- **Conditional support asset:** `prostudio_image_generate` only when the user approves a still environment asset that is genuinely needed before the final video generation. The finished Foldscape result must remain one video clip.\n- This declaration does not activate tools. Use only the tools required by the chosen workflow. If a required tool is absent from the current scoped catalog and `load_prostudio_tools` is available, call it with the missing outcome before claiming the capability is unavailable. If retrieval still exposes no matching tool, report that limitation. A model-catalog lookup alone is never completion for a requested Foldscape shot.\n\nCreate one high-end Hollywood feature-film VFX shot in which a recognizable subject walks through the scene and looks around while the surrounding world lifts, bends, and folds inward into a monumental gravity-defying enclosure. Aim for premium production design, photoreal integration, deliberate cinematography, physically coherent lighting, and finished theatrical-scale visual effects rather than a social-media filter or technical demo. The environment performs the transformation; the walking character must remain anatomically coherent and must not stretch, morph, grow, rotate with the world, or become the transition surface.\n\nUse semantic skill selection only. Do not add deterministic routing, keyword matching, catalog changes, or UI changes as part of this workflow.\n\n## Minimum input and asset policy\n\n- One clear reference of the intended character or subject is sufficient to begin. Treat it as the identity reference and preserve face, body proportions, hair, wardrobe, accessories, and defining features.\n- An environment image or video is optional. When supplied, preserve its recognizable geography, architecture, lighting direction, horizon, and material character while folding it. When absent, derive the setting from the Brief or the user's request and generate it as part of the video.\n- Do not demand a separate clean plate, depth map, city reference, or style frame when a live reference-to-video model can create the requested result from the character reference alone.\n- If both setting and visual mood remain undefined, ask one concise question that resolves them together. Otherwise infer non-critical details from the user's request and, when present, the optional Brief. Never create or require a Brief to run Foldscape.\n- Load `media-inspection` and inspect every selected reference before planning. Filenames, thumbnails, prompts, and summaries are not evidence that identity, framing, or image quality is suitable.\n- Reuse user-designated assets through `reference-management`. Never replace the selected identity without the user's choice.\n\n## Establish the effect\n\n- The subject must walk continuously along a readable path and actively observe the transformation with changing eyelines, head turns, brief glances upward and sideways, and believable balance reactions. Do not leave the character standing still or posing passively.\n- Separate the scene into a **protected walking corridor** and a **folding environment zone**. The ground immediately under the subject's current and upcoming footsteps remains level and geometrically stable enough to support the walk; the surrounding ground, streets, buildings, terrain, sky-facing structures, and distant landscape perform the lift and fold.\n- Preserve a convincing gait: alternating planted feet, natural weight transfer, stable body proportions, forward progression, and continuous contact shadows. The character may slow or briefly hesitate in reaction, but must keep moving unless the user explicitly requests a stop.\n- Define a readable fold origin and path. A strong default is: the far background lifts first, the deformation travels through the midground on coherent hinge lines, then the left and right environment curl inward and overhead while the subject's immediate platform stays intact.\n- Preserve material continuity. Streets bend as continuous surfaces, façades keep believable thickness, windows and structural lines follow the deformation, vegetation and loose objects respond to changing orientation, and connected geometry does not tear or duplicate without cause.\n- Preserve scale through recognizable anchors such as streets, windows, lamp posts, trees, vehicles, cliffs, or horizon lines. The final enclosure should feel massive enough to surround the subject, not like a miniature shell or decorative backdrop.\n- Treat gravity as surreal but internally coherent. The environment may rotate beyond normal physics, while loose objects, dust, vegetation, reflections, shadows, and secondary motion respond progressively to the new orientation.\n\n## Build the causal motion\n\n- Default to one continuous clip so the transformation is visibly caused rather than hidden by cuts. Use an edited structure only when the user explicitly requests it and the fold remains spatially understandable.\n- Stage the motion in five causal phases:\n  1. **Baseline:** the subject is already walking through an ordinary, readable environment.\n  2. **Lift:** the distant environment begins to rise; perspective and horizon deformation become visible before the enclosure forms.\n  3. **Propagation:** the fold travels toward the midground and side architecture with staggered structural and secondary reactions.\n  4. **Enclosure:** the lifted world curls inward and overhead around the moving subject, creating a vast gravity-defying chamber without touching or deforming the protected walking corridor.\n  5. **Settle:** motion decelerates with residual dust, sway, vibration, reflections, and shadow movement, leaving the subject readable inside the transformed world.\n- Use progressive acceleration and deceleration. The transformation must not pop into place, behave like a flat page turn, or move every structure on the same frame.\n- Keep the walking performance reactive and believable: the subject looks upward and to both sides as different parts of the world fold, briefly adjusts pace, braces through natural steps, and lets hair and clothing respond to air displacement. The attention changes must follow the location and timing of the deformation rather than becoming random head movement.\n\n## Direct the cinematic capture\n\n- The camera must move throughout the shot and create genuine perspective change around the walking subject. It must not remain fixed directly in front of the character, preserve one frontal composition, or merely retreat at a constant distance.\n- Favor a motivated three-quarter tracking path that evolves through the shot: begin behind or beside the subject, move into a lateral or leading three-quarter view as the first fold rises, then add a restrained crane, arc, or elevation change to reveal the enclosure overhead. Preserve smooth parallax, screen direction, walk continuity, and spatial legibility rather than orbiting decoratively.\n- Keep the walking subject prominent and the fold origin readable; do not lock the character in place through artificial treadmill motion. Camera acceleration, height, angle, and distance should develop with the environmental escalation while remaining controlled enough for feature-film coverage.\n- Avoid a lens-axis roll, full camera flip, frantic orbit, whip pan, crash zoom, teleport, or random drone movement. The world bends around the camera; the illusion must not be reduced to rotating the camera or horizon.\n- Specify camera height and angle, lens focal length and character, focus distance, depth of field, foreground occlusion, composition, aspect ratio, frame rate, motion-blur feel, exposure, color temperature, contrast, and motivated light direction.\n- A strong default visual target is an ARRI ALEXA 35 or LF-style cinematic image, natural skin tones, controlled highlight roll-off, detailed shadows, restrained film grain, realistic motion blur, and a coherent LogC-style grade. Describe this only as capture intent; never claim that a physical camera or codec was used unless the tool reports it.\n- Keep the subject's face and silhouette readable as the environment crosses behind and above. Preserve separation with motivated rim light, atmospheric depth, selective focus, and coherent exposure rather than artificial cutout edges.\n- Let shadows, reflections, haze, and light evolve with the fold. Moving architecture must cast moving occlusion and shadow; reflective surfaces must change their environment response; volumetric depth must follow the transformed geometry.\n\n## Prompt strategy\n\n- Read the live model catalog and select a currently available reference-compatible video mode. Do not hardcode a provider or promise exact identity, geometry, or physical simulation.\n- Prepare exactly one final video generation and execute exactly one `prostudio_video_generate` action. Do not generate the phases as separate shots and do not use the Editor to assemble them.\n- Write one spatially explicit English prompt that names the bound character reference, protected walking corridor, walking path, gaze choreography, environment, fold origin, propagation path, enclosure shape, camera package, lens, movement, lighting, timing, aspect ratio, and final composition.\n- Use this structure and replace every bracketed field with concrete information:\n\n  ```text\n  ONE CONTINUOUS HIGH-END HOLLYWOOD FEATURE-FILM FOLDSCAPE VFX SHOT, premium photoreal integration and theatrical production value.\n  IDENTITY LOCK: [exact bound character reference, face, body, hair, wardrobe, accessories].\n  CHARACTER PERFORMANCE: [walking direction, pace, gait, forward progression, gaze choreography and reactions]; identity and anatomy remain stable, feet plant naturally, and the protected walking corridor never bends beneath the active footsteps.\n  ENVIRONMENT: [location, architecture or terrain, time of day, scale anchors, material character].\n  0–[time]: BASELINE — [readable composition and ordinary environmental motion].\n  [time]–[time]: LIFT — [distant fold origin, horizon deformation, first structural and secondary response].\n  [time]–[time]: PROPAGATION — [coherent hinge lines, staggered midground and side movement, material continuity].\n  [time]–[time]: ENCLOSURE — [world curls inward and overhead around the walking subject, monumental final geometry, subject looks toward the active fold while continuing forward].\n  [time]–END: SETTLE — [deceleration, residual dust, sway, vibration, moving shadows and reflections].\n  CAMERA: continuously moving [evolving three-quarter tracking path, real perspective change, parallax, height and angle progression]; never fixed directly in front of the subject and never maintaining one constant frontal distance; [format intent, lens, focus and depth of field, frame rate and motion blur].\n  LIGHT AND GRADE: [source direction, exposure continuity, skin tone, highlight roll-off, contrast, atmosphere, color intent].\n  CONTINUITY: one identity, one outfit, one location, one continuous camera path, one physically continuous fold.\n  ```\n- Include positive behavior before negative constraints. State exactly which surfaces lift, where each hinge line travels, how the fold surrounds the subject, and how secondary physics confirms the motion.\n- Explicitly forbid a stationary or passive subject, frozen posing, treadmill walking, foot sliding, gait loops that do not advance, a locked frontal camera, fixed head-on framing, constant-distance backward tracking, static tripod coverage, subject deformation, identity drift, duplicate subjects, rotating ground beneath active footsteps, flat-cardboard layers, texture stretching, melting buildings, duplicated architecture, intersecting geometry, sudden scene replacement, portals, kaleidoscopic patterns, generic camera rolls, unexplained destruction, hard cuts, and a static background with camera-only movement.\n- Prepare the exact generation, report the returned model, duration, aspect ratio, audio behavior, and credit charge, then stop for the host's required confirmation. After approval, execute the prepared action without changing its arguments, poll the durable job, and inspect the completed video before calling the effect successful.\n\n## Preflight checklist\n\nBefore generation, verify that:\n\n- one exact identity reference is bound and visually suitable;\n- the subject, wardrobe, walking direction, pace, gaze choreography, and protected walking corridor are explicit;\n- the setting is either supplied or concretely defined for generation;\n- fold origin, direction, hinge lines, propagation order, and final enclosure are spatially explicit;\n- the baseline, lift, propagation, enclosure, and settle phases fit the requested duration;\n- the effect is one continuous environment transformation, not a camera roll or hidden scene transition;\n- camera movement changes perspective, height, angle, or relative position throughout the shot, keeps the subject readable, never settles into fixed frontal coverage, and does not imitate the fold by rotating the frame;\n- geometry, materials, shadows, reflections, atmospheric depth, and loose objects all respond coherently;\n- the prompt contains one final generation, no placeholders, no separate phase renders, and no Editor assembly;\n- the live model supports the chosen reference binding, duration, aspect ratio, and requested audio behavior.\n\n## Review and revision criteria\n\n- **Pass:** the selected subject remains recognizable and walks continuously with natural forward progression, planted footsteps, weight transfer, and gaze reactions aimed toward the parts of the world that are actively folding.\n- **Pass:** the shot has premium feature-film production value and the camera follows an evolving cinematic path with clear parallax and perspective change rather than a fixed frontal view.\n- **Pass:** the background visibly lifts from a defined origin, the deformation propagates through continuous geometry, and the world folds inward and overhead into a massive enclosure.\n- **Pass:** the transformation reads through perspective, scale anchors, parallax, material thickness, moving occlusion, shadows, reflections, atmosphere, and secondary physics rather than a rotated camera.\n- **Pass:** motion follows baseline → lift → propagation → enclosure → settle with readable acceleration, staggered reactions, and a resolved final composition.\n- **Pass:** the result is one coherent cinematic clip and one location transformed in place, with consistent identity, wardrobe, lighting, and camera logic.\n- **Fail:** the subject stands still, poses passively, walks in place, slides, repeats an artificial gait loop, looks randomly away from the active fold, or has face, body, clothing, or foot-contact geometry bend, grow, duplicate, rotate, or merge with the environment.\n- **Fail:** the camera stays fixed in front, retreats at one constant distance, uses static tripod coverage, or lacks meaningful parallax and changing perspective around the walking subject.\n- **Fail:** the result is mainly a camera roll, horizon rotation, zoom, portal, scene replacement, flat page curl, or layered 2.5D parallax instead of a volumetric environmental fold.\n- **Fail:** buildings melt, surfaces stretch without thickness, roads disconnect, architecture duplicates, geometry clips through the subject, or the enclosure becomes a miniature shell.\n- **Fail:** the whole environment moves simultaneously, the fold pops into place, the final enclosure never forms, or the clip ends before the motion settles.\n- **Fail:** the reference remains a frozen cutout, only the background moves while the subject does not walk and react, or the subject loses visual separation.\n- For a localized failure, revise only the responsible walking path, gait, gaze cue, protected corridor, hinge path, depth cue, secondary physics, timing, or camera constraint, then regenerate the complete single clip. Do not patch the result by hiding the failure with a cut or by generating separate phases.\n"},"changes":[],"summary":"First saved snapshot. No earlier version is available for comparison.","summary_kind":"deterministic","summary_metadata":{}}