← WidemileCONTENT HISTORY
WHAT CHANGED
Update to Widemile
Snapshot Oct 24, 2026 · 13:07 UTC · version 1.1.0
Collection source: downloaded plugin package.
First saved snapshot
No earlier snapshot is available to establish a change.
Save Widemile and check future changes in Following.
Up to 3 plugins free · no card required.
Compare saved observations
Download comparison JSONFull technical diff · 0 changed fields
Full snapshot data
{
"description": "Create one continuous cinematic descent from Earth orbit through atmosphere, clouds, city, and street level into one supplied endpoint image or video; use for space zooms, orbital dives, planet-to-street reveals, extreme scale transitions, zoom dallo spazio, discesa orbitale, zoom espacial, or satellite-to-ground shots, not for ordinary digital zooms, unrelated aerial footage, or multi-clip travel montages.",
"included_files": [],
"name": "vfx-space-zoom",
"skill_md_contents": "---\nname: vfx-space-zoom\ndescription: Create one continuous cinematic descent from Earth orbit through atmosphere, clouds, city, and street level into one supplied endpoint image or video; use for space zooms, orbital dives, planet-to-street reveals, extreme scale transitions, zoom dallo spazio, discesa orbitale, zoom espacial, or satellite-to-ground shots, not for ordinary digital zooms, unrelated aerial footage, or multi-clip travel montages.\n---\n\n# Space Zoom VFX\n\n## Tool requirements\n\n- **Core evidence and endpoint binding:** `prostudio_brief_get`, `prostudio_workspace_get`, `prostudio_references_list`, `prostudio_references_resolve`, `get_prostudio_media_catalog`, and `inspect_prostudio_media`.\n- **Final Space Zoom 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 missing support still that is genuinely necessary. The supplied image or video remains the destination authority, and the finished result must remain one video clip.\n- This declaration does not activate tools. Use only the tools required by the selected 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 Space Zoom.\n\nCreate one uninterrupted high-end cinematic move in which the camera begins in Earth orbit, dives through the atmosphere and cloud layers, resolves the city grid and neighborhood, enters the street canyon, and lands precisely in the supplied destination shot. The effect is a physical scale traversal with continuously increasing parallax and detail, not a digital push-in, a sequence of unrelated aerial images, or several clips hidden behind clouds.\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 endpoint policy\n\n- Exactly one user-selected endpoint asset is sufficient: either one image or one video. Treat it as the destination authority for final composition, subject identity, architecture, street geometry, time of day, light direction, color, and visible action.\n- Load `media-inspection` and inspect the selected endpoint before planning. Filenames, thumbnails, prompts, and summaries are not evidence that its framing, content, resolution, motion, or lighting is suitable.\n- Reuse the exact selected asset through `reference-management`. Do not replace it with a newly generated approximation unless the user explicitly chooses another destination.\n- Do not require a separate satellite image, cloud plate, aerial plate, city map, clean plate, or start frame when a compatible live model can generate the scale traversal from the endpoint alone.\n- If the endpoint clearly identifies a real location, preserve recognizable geography and approach from the correct city context as far as the available model can support. If it does not, create a plausible macro geography that converges coherently on the visible destination; do not invent a famous landmark or claim geographic accuracy.\n- Ask for a location only when exact real-world geography is central to the request and cannot be determined from the endpoint or Brief. Otherwise derive non-critical macro detail and continue.\n\n## Interpret image and video endpoints\n\n- **Image endpoint:** the still becomes the final street-level composition, but it must come alive before arrival completes. Add subtle subject motion when people are present, traffic and pedestrian motion when appropriate, moving foliage, atmospheric drift, reflections, practical-light behavior, and natural handheld or stabilized residual motion. Never finish on a frozen photograph or a flat 2.5D card.\n- **Video endpoint:** preserve its subject, action, camera direction, cadence, and environment. Design the descent to merge into a compatible moment near the beginning of the supplied motion, then continue that motion long enough for the destination to register. Do not reverse, freeze, restart, or arbitrarily retime the source action.\n- In both modes, match the final lens perspective, camera height, horizon, vanishing point, exposure, weather, light direction, and color before the orbital descent resolves into the endpoint.\n\n## Design the scale traversal\n\n- Default to one continuous shot with no editorial cuts. The visible journey must pass through all scale bands in causal order:\n 1. **Orbit:** curved Earth, thin atmospheric limb, planetary light, and a clear target region.\n 2. **Atmospheric entry:** controlled acceleration, atmospheric haze, large cloud systems, and increasing surface detail.\n 3. **Regional approach:** landform or coastline continuity gives way to roads, districts, and a legible city grid.\n 4. **Urban descent:** blocks, rooftops, streets, traffic, and architecture gain depth and speed with strong but coherent parallax.\n 5. **Street arrival:** the move decelerates, aligns to the endpoint lens and horizon, clears the final foreground occlusion, and resolves into the supplied image or video.\n- Lock that order as **Orbit → Atmospheric entry → Regional approach → Urban descent → Street arrival**. Never reverse, reshuffle, or repeat a scale band.\n- Give each band enough screen time to read while maintaining relentless forward momentum. Compress uneventful altitude ranges instead of skipping them through a cut or teleport.\n- Use progressive acceleration from orbit, maximum controlled velocity through the macro-to-city transition, then a smooth deceleration through rooftops and street architecture. Avoid an abrupt stop at the endpoint.\n- Maintain one target lock. The intended destination remains on a coherent approach vector; coastlines, street grids, landmarks, and building geometry must not reconfigure randomly between scales.\n- Clouds may add depth and speed, but they must remain translucent, layered atmospheric geometry. Never use a full-frame cloud, white flash, darkness, lens flare, smoke wall, or motion-blur wipe to hide a scene replacement.\n\n## Direct the cinematic capture\n\n- Treat the move as a camera physically traversing space, atmosphere, air, and the built environment. Perspective, field of view, motion blur, haze, parallax, and relative speed must evolve with altitude and proximity.\n- Keep the path dynamic but controlled: a slight orbital arc can straighten into a steep dive, then transition into a precise oblique urban approach before leveling into the destination camera height. Do not spin, barrel-roll, ricochet, orbit the city, or wander away from the target.\n- Specify camera and lens behavior for every scale band. Begin with a very wide orbital field of view, narrow or normalize through the regional approach, and match the endpoint lens exactly for the street arrival. Avoid fisheye distortion, impossible focal-length jumps, or a constant digital zoom.\n- Target premium feature-film VFX: photoreal Earth and atmosphere, detailed cloud volumes, large-scale production design, physically coherent daylight or night lighting, controlled highlight roll-off, rich shadow detail, realistic motion blur, restrained grain, and seamless final compositing.\n- An ARRI ALEXA 35 or LF-style image can be used as the ground-level capture intent when appropriate. Describe it as a visual target only; never claim a physical camera, lens, codec, or color pipeline unless the tool reports it.\n- Preserve readable foreground occlusion near street level: rooftops, building edges, signs, trees, vehicles, or architectural elements should cross with correct parallax and reveal the endpoint naturally rather than clipping or dissolving.\n\n## Prompt strategy\n\n- Read the live video-model catalog and select a currently available mode that supports the supplied image or video endpoint. Do not hardcode a provider or promise exact geography, identity, or frame-perfect source continuation.\n- Prepare exactly one final video generation and execute exactly one `prostudio_video_generate` action. Do not generate orbital, aerial, and ground phases separately, and do not use the Editor to assemble them.\n- Write one spatially explicit English prompt that identifies the exact bound endpoint, endpoint type, target composition, scale bands, approach vector, timing, camera and lens evolution, atmospheric depth, geographic continuity, deceleration, endpoint animation or continued source action, aspect ratio, lighting, grade, and audio intent.\n- Use this structure and replace every bracketed field with concrete information:\n\n ```text\n ONE UNBROKEN HIGH-END CINEMATIC SPACE-TO-STREET CAMERA DESCENT, no cuts and no hidden scene replacement.\n DESTINATION LOCK: [exact bound image or video], [subject/action], [street geometry], [final camera height, lens, horizon, vanishing point, light, weather, color].\n ORBIT [start–time]: [Earth limb, target region, orbital arc, planetary light, wide field of view].\n ATMOSPHERIC ENTRY [time–time]: [acceleration, haze, layered cloud volumes, growing terrain detail, target remains locked].\n REGIONAL APPROACH [time–time]: [continuous landform/coastline/road geography, districts and city grid resolve, evolving lens and parallax].\n URBAN DESCENT [time–time]: [oblique approach, rooftops and streets, increasing relative speed, foreground occlusion, smooth deceleration].\n STREET ARRIVAL [time–end]: [precise endpoint match, final lens and camera height, continued image animation or video action, controlled settle].\n CAMERA CONTINUITY: one physical trajectory, continuous forward motion, evolving field of view and motion blur, no roll, teleport, cut, dissolve, flash, cloud wipe, or digital zoom substitution.\n ```\n- Include positive continuity cues before negative constraints. State how planetary, regional, urban, and street geometry progressively reveal the same destination.\n- Explicitly forbid hard cuts, dissolves, morphing maps, unrelated cities, duplicated buildings, changing road layouts, fake satellite UI, map labels, text overlays, cloud-cover edits, white flashes, portals, tunnels, barrel rolls, planet deformation, flat Earth appearance, miniature city appearance, texture boiling, fisheye warping, digital-only zoom, clipping through solid structures, and a frozen final image.\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- exactly one image or video is explicitly bound as the destination authority;\n- the endpoint has been inspected and its final composition, subject or action, lighting, horizon, and lens character are understood;\n- the orbital target, approach direction, city context, street path, and endpoint alignment form one coherent route;\n- orbit, atmosphere, regional approach, urban descent, and street arrival have explicit timing and no omitted scale band;\n- motion accelerates and decelerates smoothly, with continuously increasing detail, speed cues, depth, and parallax;\n- clouds add atmosphere without hiding a cut or unrelated scene replacement;\n- an image endpoint receives believable living motion and a video endpoint preserves and continues its source action;\n- the prompt describes one generation, one clip, and one physical camera path with no placeholders or Editor assembly;\n- the live model supports the selected endpoint type, duration, aspect ratio, and requested audio behavior.\n\n## Review and revision criteria\n\n- **Pass:** one unmistakably continuous camera path travels from Earth orbit through atmosphere, regional geography, city grid, rooftops, and street level.\n- **Pass:** the same target remains spatially coherent across scale bands and the move resolves precisely into the selected image or video composition.\n- **Pass:** perspective, parallax, field of view, motion blur, haze, detail frequency, and relative speed evolve naturally with altitude.\n- **Pass:** clouds and foreground architecture create depth without masking a cut, and the street arrival decelerates into a readable cinematic reveal.\n- **Pass:** a still endpoint is visibly alive; a video endpoint continues with matching direction, cadence, action, and camera logic.\n- **Fail:** the result is a digital zoom, a Google-Earth-like texture push, a montage, or multiple shots joined behind clouds, blur, darkness, flares, or flashes.\n- **Fail:** geography changes arbitrarily, the target drifts, city blocks reconfigure, buildings duplicate, the camera teleports, or the endpoint is only approximately recreated.\n- **Fail:** the planet looks flat or deformed, the city reads as a miniature, cloud layers are planar, or the camera clips through solid structures.\n- **Fail:** the camera spins, barrel-rolls, pauses mid-descent, stops abruptly, misses the destination lens and horizon, or lands on a frozen frame.\n- For a localized failure, revise only the responsible scale band, route, lens transition, target lock, cloud depth, deceleration, or endpoint binding, then regenerate the complete single clip. Do not hide the failure with a transition or generate separate altitude phases.\n"
}SHA-256 of public snapshot: 3800682ec234dd6e5e66804eb51168e68134eddc7beb5f07415c972028afae88