Loova

View saved version →

---
name: loova-ad-resizer
description: "Adapt an existing product or ad image to requested dimensions and aspect ratios with Loova-compatible editing or deterministic resizing. Preserve product proportions, copy, and required content through containment, cropping, extension, or reflow."
metadata:
  version: "2.1.4"
  brand: "Loova"
  category: "image-creation"
---

# Ad Resizer

Adapt one creative to different placements without stretching the product or sacrificing essential information.

## Choose a method

Require the source, target dimensions or ratios, and protected content. If only a destination is supplied, use a provided template or verify its current dimensions. Carry the same source and product constraints across all requested outputs.

| Method | Use when | Preserve |
|---|---|---|
| Contain | The complete image or copy must remain visible | Proportions, with deliberate margins if needed |
| Crop | All essential content fits within a safe crop | Product, price, logo, and required disclosures |
| Extend | Additional background is needed and generation is allowed | Product identity and existing copy |
| Reflow | Elements can be rearranged for a new layout | Meaning, hierarchy, and exact content |

Prefer reflow when editable source layers exist. Do not pretend flattened text can be moved losslessly. Explain any reconstruction needed to expose covered areas or rebuild text.

## Workflow

1. Mark product, logo, price, headings, disclosures, and non-croppable regions. Identify background that can be cropped or extended.
2. Calculate each target separately. For source W x H and target w x h, contain uses min(w/W, h/H); cover-and-crop uses max(w/W, h/H). These formulas do not make a centered crop safe automatically.
3. If important content cannot fit readably, use permitted margins or reflow. Extend only the background; do not invent hidden product faces, accessories, or new claims.
4. Read [Loova execution](references/execution.md). Use a supported deterministic resize/composite path or Loova generative extension. If exact export dimensions are unsupported, report actual dimensions rather than claiming an exact-size delivery.
5. Inspect every target independently. Check required text, real pixel dimensions, product geometry, texture continuity, and any UI-safe area.

## Delivery

Return a target-to-file map with actual dimensions and the chosen method. Preserve the source. Identify unmet targets separately; do not label one square file as several different exports.

Referenced files: 2 — file references are metadata, not package contents.

View saved version →

---
name: loova-ai-image-editor
description: "Edit a supplied image with Loova, changing specified objects, backgrounds, details, or visual style while preserving the agreed subject and composition. Use for image edits; not for creating an unrelated image or only resizing a layout."
metadata:
  version: "2.1.4"
  brand: "Loova"
  category: "image-creation"
---

# AI Image Editor

Make a bounded change to an existing image. Identify the edit area and protected content before choosing a method.

## Define the edit

Require the actual source image. Identify the target object or region, desired change, and protected subject, text, and layout. If a phrase such as "make this red" has several plausible targets, clarify the target rather than redesigning the image.

Use local editing for object removal, recoloring, and small defects; foreground protection and compositing for background changes; or a global style transformation with identity and composition constraints for restyling.

## Workflow

1. Inspect boundaries, occlusion, shadows, and reflections. Removing an object may require filling hidden background; a plausible reconstruction is not proof of the original scene.
2. Record what may change and what must remain. A watercolor restyle may change brushwork and material rendering without changing identity, product structure, or object count.
3. Use the smallest effective editing region. Update affected shadows and reflections together with the object. For a background replacement, distinguish surface color from environmental color spill.
4. Read [Loova execution](references/execution.md). Use a tool that accepts the required source image and edit controls. Protect exact pixels through masking or compositing when available; do not claim pixel preservation for a full-image regeneration.
5. Compare source and output. Inspect the intended region and protected areas, including text, logos, transparent edges, hair, fine openings, perspective, and contact shadows.

## Repair and delivery

Repair seams at the seam, lighting at the affected light boundary, and identity drift from the original source. Do not keep repainting a drifted result. If the available tools cannot meet a preservation requirement, explain the specific limitation and offer a suitable alternative.

Return the edited image and one sentence describing the change. Keep the source and identify meaningful differences between requested variants. If execution is unavailable, label the output as an edit plan, not a completed edit.

Referenced files: 2 — file references are metadata, not package contents.

View saved version →

---
name: loova-character-swap
description: "Replace a specified person or character in source footage using Loova video editing, retaining the agreed scene, motion, camera, timing, and audio. Use for identity replacement in an existing video; not for performing reference motion in a new scene."
metadata:
  version: "2.1.4"
  brand: "Loova"
  category: "video-creation"
---

# Character Swap

Replace a defined character in existing footage. The source supplies timing, movement, camera, and scene; the target image supplies identity.

## Required sources and scope

Require accessible source video and a target identity reference. Identify the target when several people are present. Clarify face-only, whole-person, clothing, or fictional-character scope when material to the result. Do not interpret a character change as permission to change voice or all styling.

Keep source audio unless the user requests otherwise. Retain clothing and prop relationships unless a full-character replacement requires a different silhouette or the user specifies another outfit. The target reference's pose and background are not automatically part of the edit.

## Workflow

1. Inspect cuts, turns, exits, rapid movement, occlusions, held objects, and interactions. Do not judge feasibility only from the opening frame.
2. Define the target and protected areas. Preserve contact points, foot placement, occlusion order, other people, background, and relevant shadows. Do not draw a replacement in front of objects that should cover it.
3. Separate source-preservation constraints from identity changes. Face-only replacement protects hair, clothing, and body. Whole-person replacement may change the silhouette without changing the original action's rhythm.
4. Read [Loova execution](references/execution.md). Require actual source-video editing or identity-replacement support. If only similar-action generation is available, offer a regeneration alternative without calling it a source-video swap.
5. For longer footage, use existing cuts or suitable segment boundaries with a consistent time map and identity reference. Check segment joins for duplicates and audio drift.
6. Reattach original audio only if the time map remains aligned; otherwise correct timing first. Treat an audio change as a separate requested edit.

## Quality and delivery

Check identity through front, side, back, and occlusion recovery. Inspect protected scenery and bystanders, held objects, crossing limbs, feet, edge blending, and shadows. Compare actual duration, cuts, frame rate, action beats, and sound sync.

Repair the affected segment or narrow the edit scope when necessary; do not change the source action to conceal a failure. Promise pixel-level preservation only when supported by the real masking or compositing path. Deliver the video with a concise scope summary and any unmet shot or synchronization requirement.

Referenced files: 2 — file references are metadata, not package contents.

View saved version →

---
name: loova-mimic-motion
description: "Make a target character perform movement from a reference video with Loova, separating identity, body motion, camera, and environment. Use for reference-motion transfer or motion plans; use source-video identity editing when the original scene must remain unchanged."
metadata:
  version: "2.1.4"
  brand: "Loova"
  category: "video-creation"
---

# Mimic Motion

Extract recognizable movement for a target character without unintentionally copying the source person's identity, setting, or camera.

## Reference roles

Require a target character image and an accessible motion video. Inspect target anatomy, clothing, and usable joint range. Identify the performer in multi-person footage. Adapt stride and amplitude for different body proportions while preserving the motion's intent rather than forcing an incompatible skeleton.

The character image provides identity and specified appearance; the motion video provides action sequence, body phase, speed, and contact. Neither automatically provides the new scene or camera. If the user needs the original scene and camera unchanged, treat the task as source-video identity replacement instead.

## Extract the motion

1. Identify preparation, initiation, peak or contact, absorption, and recovery with visible time points and support feet. For dance, inspect weight and limb phase; for handling, inspect hands and objects; for jumps, inspect takeoff and landing.
2. Separate camera and body movement. A pan is not body translation, and a zoom is not forward movement by the character.
3. Identify hidden or ambiguous gestures rather than treating them as observed. A text description permits a text-based motion plan, not a claim of extracted trajectories.
4. Build a compact beat map: time range, pose, support or contact, direction, relative speed, and next state. Use meaningful phases rather than unnecessary frame-by-frame lists.

## Transfer and execute

1. Retarget for the character's proportions, prioritizing direction, timing, balance changes, and recognizable poses.
2. Frame the new scene so important limbs and contact points remain visible. Use a clear full-body view where useful; do not hide errors behind rapid camera movement.
3. Preserve specified rhythm. Loova's dedicated motion tool derives duration from the reference and model rules; do not pass an arbitrary duration override. If retiming is requested, explain any required supported source-editing step before creating the task.
4. Read [Loova execution](references/execution.md). Verify motion-reference, pose-control, or retargeting support. Text-to-video alone may approximate an action but is not proof of precise reference following.
5. Join segments at stable poses or compatible phases. Keep left/right direction and endpoint continuity; avoid arbitrary cuts halfway through landing or rotation.

## Quality and delivery

Check action order, limb direction, phase, support, contact, anatomy, and clothing stability. Two clips both containing dance do not establish motion correspondence. Inspect foot sliding, penetration, and recovery. Check that source camera, background, and identity have not leaked into the result.

Repair the specific phase, support, or amplitude issue and reduce unnecessary camera movement. Do not disguise errors through slow motion or occlusion. Deliver the requested video or beat plan, explaining approximation and inspection limits without claiming frame-level matching from sampled checks.

Referenced files: 2 — file references are metadata, not package contents.

View saved version →

---
name: loova-product-listing-images
description: "Create coordinated product listing images with Loova, including hero, feature, detail, lifestyle, and specification images. Use for ecommerce image sets with consistent SKU facts; not for video or a small edit to a single image."
metadata:
  version: "2.1.4"
  brand: "Loova"
  category: "image-creation"
---

# Product Listing Images

Build a listing image set in which every image answers a distinct buying question. Maintain one confirmed SKU, a coherent color system, and a clear text hierarchy.

## Establish the facts

Read product photos, SKU information, destination, and supplied dimensions, materials, accessories, functions, and evidence. Separate supplied facts, directly visible facts, and unknowns. Never infer dimensions, certifications, performance, sales, or comparison claims from appearance.

If no destination is specified, use a general listing layout without claiming platform compliance. Consult current official requirements only when the destination makes them relevant. Keep explicit output counts. For an unspecified full set, propose a useful mix of main, feature, detail, lifestyle, and specification images without inventing benefits to fill slots.

## Workflow

1. Assign each image an information goal, supporting fact, required angle, and exact copy. Separate main-image requirements from promotional secondary-image conventions.
2. Set consistent product placement, headings, backgrounds, annotations, and whitespace. Preserve real packaging copy where possible; compose additional text separately when the tools support it.
3. Generate the required product scenes and assemble accurate copy and specifications. Read [Loova execution](references/execution.md) before media operations. If text layout or a required module is unsupported, deliver completed assets and identify the remaining composition step.
4. Check the actual reading size, especially on mobile. Verify every number, unit, annotation endpoint, and product feature.
5. Keep sets within the user's chosen colorways and variants. Additional colorways are acceptable when requested; identify simulated options rather than presenting unverified variants as products available for sale.

## Quality and delivery

Check consistent color, accessories, quantities, ports, and product version across the set. Props must not appear to be included accessories. Use real detail references for close-ups; label explanatory diagrams as diagrams rather than claiming they reveal hidden construction. Resolve conflicting facts instead of choosing a convenient value.

Fix a text layout defect locally; fix product identity before rebuilding a set. Deliver in the requested order with the actual files and a short image map. State missing facts or unfinished modules. A completed image set does not imply marketplace approval.

Referenced files: 2 — file references are metadata, not package contents.

View saved version →

---
name: loova-product-photoshoot
description: "Create studio, lifestyle, or campaign images for a specific product using Loova. Use for new product photography from product references or a concept brief; use an editing or listing workflow for local retouching or a product listing set."
metadata:
  version: "2.1.4"
  brand: "Loova"
  category: "image-creation"
---

# Product Photoshoot

Create product images with consistent identity, credible materials, and intentional photography. Deliver images, a plan, or prompts according to the user's request.

## Inputs and defaults

Use real product references when available. Extract visible silhouette, proportions, color, structure, materials, logo, and packaging copy. With text alone, create a concept product and identify it as such. Reuse context rather than asking for information already supplied. For an unspecified single image, use one square image as a working default if the selected model supports it.

Choose studio for a clean product presentation, lifestyle for a believable use setting, or campaign for a specific brand message with deliberate copy space. A request to change only an existing background belongs to image editing. A multi-module shopping page belongs to listing images; those skills need not be installed to explain the distinction.

## Workflow

1. Lock visible product features. Separate permissible changes in photography from unknown rear surfaces, accessories, and functions. Do not invent concealed product details.
2. Define the relationship between product, camera height, viewing angle, framing, key light, fill, background distance, and supporting surface. Use controlled broad reflections for reflective materials; consider transmission and edge light for transparent objects.
3. For studio work, prioritize silhouette and material. For lifestyle work, choose a plausible use moment before adding props. For campaign work, establish the message and copy area before arranging the product. Do not add splashes, levitation, or people by default.
4. Treat the product reference as identity and any scene reference as environment or style. Read [Loova execution](references/execution.md) before creating media and translate the intent into supported tool fields.
5. For a set, keep product identity and lighting logic consistent. Change the requested scene or camera variables without introducing unnecessary variants.

## Request brief

Include only relevant fields: intended use and aspect ratio; product reference and locked features; studio/lifestyle/campaign mode; orientation and framing; light and supporting surface; purposeful props; copy space; permitted changes.

## Quality and delivery

Check caps, buttons, ports, proportions, packaging, and logo. Do not mark unreadable text as verified. Look for floating products, inconsistent reflections, plastic-looking metal or glass, and depth of field hiding essential details. Confirm realistic product scale and contact with hands or surfaces.

Repair identity problems through the product constraints, material problems through lighting and reflection, and unclear compositions by simplifying props. Prefer a viable local repair over regenerating a successful composition. Deliver the requested count with recognizable filenames and a brief result description; include prompts or production notes only when useful or requested.

Referenced files: 2 — file references are metadata, not package contents.

View saved version →

---
name: loova-product-video-maker
description: "Turn product images into showcase videos with Loova, coordinating camera motion, product motion, and shot continuity. Use for image-to-product-video work or storyboards; not for presenter ads, reference-ad recreation, or character motion transfer."
metadata:
  version: "2.1.4"
  brand: "Loova"
  category: "video-creation"
---

# Product Video Maker

Give a static product image a clear showcase purpose. Motion should reveal silhouette, material, structure, or a supported use, rather than causing arbitrary floating or deformation.

## Inputs and motion choices

Inspect product images, available angles, use, and requested duration, format, and sound. Confirm matching SKUs before assigning references to full views, details, and scenes. A front view is not evidence of a real back or internal structure.

If a generic short showcase has no duration, propose approximately ten seconds, subject to the selected tool's supported durations. Preserve explicit timing and silence requirements. Do not add a presenter or voiceover by default.

Choose a hero shot for silhouette and material, a detail shot for documented features, a context shot for plausible use, or a purposeful multi-shot sequence. Prefer gentle camera travel or limited rotation when available references cannot support revealing unknown surfaces.

## Workflow

1. Identify the main feature and locked structure. Do not invent waterproofing, disassembly, or other functional demonstrations from appearance.
2. Assign motion to the camera, product, or environment. Simple showcases usually need one main source of motion; combine them only when useful and supported.
3. Define the start and end state. Obtain more angles or choose limited movement before exposing an unknown side. Do not call a partial turn a verified 360-degree reconstruction.
4. Write shot cards with duration, reference roles, framing, main motion, amplitude, speed, endpoint, and audio. For multiple shots, establish shared product references and keyframes when supported.
5. Read [Loova execution](references/execution.md). Confirm image references, frame controls, duration, and assembly support. Prompt wording is not proof that a model locks those controls precisely.

## Quality and delivery

Inspect the opening, middle, ending, and cuts for swelling, melting, changing parts, and unstable logos. Check reflections, contact shadows, background parallax, and hand contact. Reuse verified product states. Reduce excessive motion when it causes identity drift rather than repeatedly regenerating a broken subject.

Return actual duration, aspect ratio, and audio status with the video. Distinguish independent clips from a fully assembled film. For an eight-second perfume push-in, avoid adding cap removal, spraying, or an unverified rear view.

Referenced files: 2 — file references are metadata, not package contents.

View saved version →

---
name: loova-thumbnail-cover-maker
description: "Create or revise video thumbnails and social covers with Loova, using a clear subject, accurate copy, and small-screen readability. Use for cover images or cover concepts; not for generating the video itself."
metadata:
  version: "2.1.4"
  brand: "Loova"
  category: "image-creation"
---

# Thumbnail & Cover Maker

Design a cover that communicates at a small size. Build visual interest from the real content rather than invented events, exaggerated outcomes, or unsupported performance promises.

## Inputs and framing

Identify the topic, audience, placement, and locked copy. When footage is available, choose a credible visual anchor from its content. Use supplied person and brand references for identity. Do not rewrite approved text without permission.

For an otherwise unspecified video cover, 16:9 landscape or 9:16 vertical can be a working layout choice. Verify current destination templates or UI overlays when they matter; do not assume all platform covers share the same fixed dimensions.

## Workflow

1. Choose one visual proposition: a result, comparison, or question. Establish a reading order between subject, background, and text rather than combining unrelated hooks.
2. Use topic-relevant subjects and expressions. Reference covers may guide readability or color relationships without supplying unrelated people, branding, or exact copy.
3. Keep text concise unless the user requests longer copy. Preserve all locked words and adapt the layout rather than silently deleting them.
4. Separate the subject with light, edge contrast, whitespace, and appropriate text backing. Avoid decorative arrows, circles, and exaggerated expressions unless useful to the brief.
5. Read [Loova execution](references/execution.md). Use supported text rendering or composite exact text where available, then verify every word.

## Quality and delivery

Check landscape work around 320 pixels wide and vertical work around 180 pixels wide as readability tests, not platform export specifications. Confirm the subject and core words remain recognizable without zooming; check spelling and protected UI areas.

For requested alternatives, vary a meaningful concept or layout choice. Do not create unrequested paid variants. Repair readability through simpler backgrounds or text layout, and identity through reference constraints. Return the requested covers with actual dimensions and any unfinished text composition. Do not promise untested click-through or conversion gains.

Referenced files: 2 — file references are metadata, not package contents.

View saved version →

---
name: loova-ugc-ad-creator
description: "Create presenter-led product ads with Loova, coordinating a natural script, product demonstration, speech, and supporting shots. Use for UGC-style ads and their scripts; not for a silent product-only showcase or identity replacement."
metadata:
  version: "2.1.4"
  brand: "Loova"
  category: "video-creation"
---

# UGC Ad Creator

Create a product ad with a coherent presenter, natural speech, and observable product value. Do not fabricate personal experience or testimonials to make the presentation sound authentic.

## Inputs and claims

Use product facts, audience, claims, presenter source, language, duration, and CTA. Preserve supplied scripts and presenter references. If no presenter is specified, a fictional adult creator can be proposed without implying a real customer endorsement.

Distinguish documented user experience, brand facts, and newly drafted ad copy. A statement such as "after using it for a month" needs supporting experience material; otherwise write an explanation or clearly framed demonstration. First-person language alone does not make a claim factual.

## Workflow

1. Select a main supported claim and visible demonstration. Write natural speech before assigning gestures. Reword, omit, or seek support for claims without evidence.
2. Break the actual script into speaking segments, product appearances, and supporting shots. Estimate only for planning; verify duration against generated speech rather than asserting lip sync from a reading-speed formula.
3. Separate presenter and product identity constraints. A product reference does not supply a person, and a person reference does not automatically supply clothing or background unless requested.
4. Read [Loova execution](references/execution.md). Verify a feasible route: native audiovisual generation, voice plus lip sync, or presenter shots with product inserts. Do not imply those capabilities exist just because image or video generation is connected.
5. Establish critical holding and contact poses before complex demonstrations. Prefer intelligible shots to unsupported hand changes, assembly steps, and transitions in one generation.
6. Assemble around actual speech length. If locked copy exceeds the requested duration, explain the conflict and offer a timing choice; do not silently delete words. Add captions when requested or included in the agreed delivery.

## Quality and delivery

Check hand anatomy, grip, orientation, and correspondence between gestures and speech. Confirm stable presenter appearance and voice, audible wording, complete endings, and actual lip sync. Ensure the visible demo supports the claims and that performance, sales, or testimonial statements are not invented.

Return a script for script-only requests or the assembled ad for production requests. State duration, audio, caption status, and incomplete assembly or lip-sync work. No media call is needed for a plan-only request.

Referenced files: 2 — file references are metadata, not package contents.

View saved version →

---
name: loova-viral-video-clone
description: "Adapt a reference video into a new product video with Loova, preserving the requested pacing and shot structure while using the user's product. Use for reference-led ad recreation; not for replacing only a person in the original footage."
metadata:
  version: "2.1.4"
  brand: "Loova"
  category: "video-creation"
---

# Viral Video Clone

Translate a reference video's creative structure into a video for the user's product. "Clone" describes reference-led recreation, not a guarantee of identical frames or commercial performance.

## Inputs and reference roles

Use an accessible reference video and target product information. The product image supplies identity; the video supplies the requested pacing, shot functions, camera, or editing reference. If the host cannot inspect it but Loova can, use the project's server-side analysis and attribute its source. If neither can inspect it, request an accessible copy or provide a labeled description-based plan without invented timecodes.

Establish which elements the user wants to follow. Do not import the original brand, person, distinctive copy, or soundtrack by default. Respect explicitly requested elements within the actual assets and tool capabilities.

## Workflow

1. Record observable events with time ranges: opening, product entry, action changes, demonstrations, and ending. Separate observation from inferred marketing purpose.
2. Give key shots a function, such as revealing scale, showing operation, or illustrating a supported benefit. Map that function to something the new product can actually do; rewrite incompatible actions.
3. Build a compact adaptation map: reference segment, shot function, new product action, new scene, retained rhythm, and excluded elements. Do not infer sales or conversion performance from the reference's appearance.
4. Lock the requested duration and aspect ratio. If duration changes, retain meaningful beat order and allow each action to complete instead of compressing all shots mechanically.
5. Define continuity between shots: orientation, position, hand contact, action endpoint, and transition. Write each shot as a start state, main change, end state, and reason for the cut.
6. Read [Loova execution](references/execution.md). Prefer Loova's clone project workflow: create, query analysis/script, quote, show the complete script and estimate, then confirm only after user approval. The backend orchestrates generation; do not also generate each shot separately. Use shot-by-shot regeneration only as an agreed alternative with suitable tools, not as a claim of source-footage editing.

## Quality and delivery

For multi-shot work, reuse product references and accepted shots, then assemble only with supported tools. A request for one clip must not become an unrequested campaign package.

Check that actions belong to the new product, product state remains coherent across cuts, approved claims have factual support, and the original identity or branding has not leaked through. Inspect actual timing and audio endings. Repair the specific identity, action, or cut problem only when requested, following the selected progression mode.

Deliver the requested adaptation plan, prompts, or finished video and briefly identify the main adaptations. For example, coffee-pouring beats can become a lamp's switch-on and illumination sequence; do not make a lamp perform the reference product's physical actions.

Referenced files: 2 — file references are metadata, not package contents.