apimageorg
OrganizationAgent skills for generating UGC-style ad video, product clips and spokesperson video with APImage — from Claude, Cursor, Codex or any MCP client.
Categories
Indexed Skills (49)
ai-avatar-presenter
Design and build a reusable synthetic presenter — generate the face, save it as a brand asset, and keep it consistent across every clip. Use whenever the user wants an AI avatar, a virtual spokesperson, a recurring on-camera host, or needs a face they can legally use in ads.
amazon-listing-images
Build a complete Amazon image set — compliant main image, the six-plus gallery slots, and what each slot should do. Use whenever the user is preparing Amazon listing images, A+ content, or asks about Amazon image requirements and gallery strategy.
apparel-product-photography
Generate apparel and fashion product images — ghost mannequin, flat lay, on-model — and handle fabric, fit and colour accuracy. Use whenever the user is photographing clothing, fashion, accessories, footwear, or asks about apparel listing images or on-model shots.
aspect-ratio-strategy
Pick the right aspect ratio per platform and design safe areas so nothing important gets cropped or covered by UI. Use whenever the user is generating video or images for social platforms, mentions aspect ratio or vertical or square format, or has content getting cropped or overlaid by platform UI.
async-video-job-orchestration
Handle APImage video jobs correctly — they are always asynchronous, so poll with get_video_generation rather than regenerating, and use webhooks at volume. Use whenever the user generates video with APImage, asks why a call returned no video, mentions polling or webhooks or job IDs, or is running a batch.
b-roll-generation
Generate supporting b-roll to cover cuts, add texture and carry voiceover, cheaply and in volume. Use whenever the user needs cutaway footage, filler shots, atmospheric clips, footage to cover a jump cut, or visuals to accompany narration.
background-removal-workflow
Cut products out of their background with remove_background, producing transparent PNG or WebP assets that feed every other workflow. Use whenever the user needs a transparent cutout, wants to remove a background, needs a product on white, or is preparing a product image for reuse across scenes.
background-replacement-scenes
Swap a product's background with replace_background, which relights the subject to match the new scene. Use whenever the user wants a product in a different setting, a new scene behind a product, seasonal or lifestyle variants, or a product moved from studio to context.
batch-video-production
Run video generation at volume — queueing against the concurrency cap, webhooks, budget planning and keeping a batch reconcilable. Use whenever the user is generating many clips, running a content pipeline, mentions batches or automation or scheduling, or needs a month of content.
before-after-transformation-video
Generate before-and-after transformation video that is persuasive without making claims the product cannot support. Use whenever the user wants a transformation clip, before-and-after content, a cleaning or beauty or repair demo, or a satisfying-result video.
camera-angle-variation
Generate a product from new angles using edit_image camera adjustment or reference sets, and know which angles are recoverable. Use whenever the user needs more angles of a product, wants a three-quarter or top-down or side view, or has only one product photo and needs a full listing set.
character-consistency-video
Keep the same character, product or background identical across many generations using APImage brand assets and reference images. Use whenever the user needs a recurring presenter, a consistent product across a campaign, a repeating background, or complains that the subject changes between generations.
draft-then-enhance-workflow
Iterate video cheaply on flux-3-video-draft and pay full quality only for the winner via enhance_video_draft. Use whenever the user is generating video with APImage, mentions credit cost or running out of credits, is iterating on a prompt, or is about to render variants at full quality.
etsy-listing-photos
Produce Etsy listing photos, where handmade authenticity matters and over-polished generated imagery actively hurts. Use whenever the user sells on Etsy or a handmade or craft marketplace, or asks how product imagery should differ there.
faceless-video-automation
Build faceless video content at volume — no presenter, no camera, consistent look across a channel. Use whenever the user wants faceless YouTube or TikTok content, automated content channels, a content pipeline without on-camera talent, or high-volume publishing.
flat-lay-composition
Generate overhead flat-lay product images with deliberate composition, lighting and prop discipline. Use whenever the user wants a flat lay, an overhead product shot, a top-down arrangement, or styled product imagery for social and secondary listing slots.
food-beverage-photography
Generate food and drink product imagery, including the regulatory line on depicting food accurately. Use whenever the user is photographing food, drink, packaged groceries, restaurant dishes, supplements or anything edible.
furniture-room-scenes
Place furniture and homeware in room scenes with correct scale, perspective and light. Use whenever the user is photographing furniture, homeware, rugs, lighting, appliances or anything that needs to be shown in a room.
image-model-selection
Choose between APImage's image models — flux-2-klein-9b, flux-2-pro, flux-2-max, gpt-image-2 and dall-e-3 — and the resolution to pair with each. Use whenever the user is generating images with APImage, asks which model to use, mentions credit cost on images, or is getting output that isn't good enough.
image-to-video-animation
Animate a still image into video with reference_images, keeping the exact subject rather than letting the model invent one. Use whenever the user wants to turn a photo into video, animate a product shot, mentions image-to-video, or is getting the wrong subject out of text-to-video.
inpainting-product-fixes
Repair specific regions of a product image with edit_image inpainting and erasing via masked regions, instead of regenerating the whole shot. Use whenever the user wants to remove an object, fix a blemish, repair a bad edge, erase a reflection, or change one part of an image.
instagram-reels-generation
Generate video for Instagram Reels specifically — where it differs from TikTok, feed crossover, aesthetic expectations and safe areas. Use whenever the user wants Reels content, Instagram video, Instagram ads, or asks how generated video should differ between Instagram and TikTok.
jewelry-reflective-products
Handle the hardest product category — jewellery, watches, glass, chrome and anything transparent or mirrored — where cutouts halo and reflections come back wrong. Use whenever the user is photographing jewellery, watches, glassware, chrome, mirrors, or any transparent or highly reflective product.
lifestyle-product-photography
Generate product images in real-world context — in use, in a setting, with people — rather than isolated on white. Use whenever the user wants lifestyle shots, in-context product photos, in-situ imagery, or images for ads and social rather than listings.
lip-sync-spokesperson-video
Generate talking-avatar and spokesperson video with generate_lip_sync, including the 30-second cap, the 3-credits-per-second cost and the face-image requirements. Use whenever the user wants a talking head, AI spokesperson, avatar presenter, voiceover-to-face video, or mentions lip sync.
model-wearing-product
Generate on-model product imagery with a consistent synthetic model, and handle the likeness, fit and body-representation constraints. Use whenever the user wants a model wearing or using a product, on-model apparel shots, or a recurring person across a catalogue.
multi-product-scene
Compose several products in one image — range shots, bundles, collections — with correct relative scale and no invented items. Use whenever the user wants a range shot, a product family image, a bundle or kit image, or several SKUs in one scene.
multi-scene-video-assembly
Build longer video from several short generated clips instead of asking one generation for a whole narrative. Use whenever the user wants a video longer than about 10 seconds, a multi-shot sequence, a video with several beats, or is getting morphing and drift from long single generations.
music-audio-pairing
Add music, sound design and voiceover to generated video, including the licensing rules that catch people out. Use whenever the user is adding audio to video, mentions music or sound design or trending audio or voiceover, or asks whether they can use a track in an ad.
packaging-mockup-generation
Generate packaging mockups and label visualisations, and handle the fact that models cannot render text reliably. Use whenever the user wants a packaging mockup, a label visualisation, a box or bottle or pouch render, or to see artwork applied to a product.
product-demo-video
Generate product demonstration video that keeps the real product accurate while showing it in use. Use whenever the user wants a product video, a demo clip, a product in use, Shopify or ecommerce product video, or has a product photo they want turned into video.
product-detail-macro
Generate extreme close-up detail shots showing material, texture, stitching and hardware — the quality signals customers look for. Use whenever the user needs macro shots, texture close-ups, material detail, hardware or construction detail, or wants to show product quality.
product-image-qa-review
Review generated product images before they ship, catching the artefacts and accuracy failures that pass casual inspection. Use whenever the user is about to publish product images, has a batch to check, or asks how to quality-check AI-generated product photography.
product-in-hand-shots
Generate products held in hands, which communicates scale and use — and handle the fact that hands are what generative models get wrong most. Use whenever the user wants a product held, an in-use shot, hands in frame, or a scale reference via a hand.
product-photo-batch-pipeline
Process a whole catalogue of product images — queueing against rate limits, budgeting credits, and keeping the batch reconcilable. Use whenever the user has many products to process, a catalogue to generate, mentions bulk or batch image processing, or is building an automated image pipeline.
product-photo-consistency
Keep a whole catalogue of product images visually consistent using brand assets, verbatim look clauses and locked seeds. Use whenever the user has many products to shoot, a catalogue that looks inconsistent, or wants images that read as one brand rather than many attempts.
product-photo-from-reference
Generate new product photography from a real product photo using reference_images, so the product stays accurate instead of being reinvented. Use whenever the user has a product photo and wants more shots, new angles, new scenes, or variations, or asks how to get consistent product imagery.
product-relighting
Change the lighting on an existing product photo with edit_image relighting and colour grading, without regenerating the shot. Use whenever the user wants to fix bad lighting, change light direction or mood, warm or cool an image, or match a photo to a set's look.
product-scale-reference
Communicate product size in imagery using familiar reference objects, hands and dimension overlays, to cut size-driven returns. Use whenever a product's size is unclear from a photo, the user has size-related returns or questions, or is photographing anything where scale isn't obvious.
product-upscaling-4k
Upscale product images with edit_image to meet marketplace resolution and zoom requirements. Use whenever the user has an image that's too small, needs 4K or high-resolution output, mentions upscaling or zoom requirements, or has a soft or low-resolution product photo.
seasonal-product-styling
Produce seasonal and campaign variants of product imagery from one master shot, using background replacement rather than reshoots. Use whenever the user needs Christmas or summer or back-to-school imagery, campaign variants, or seasonal refreshes of a catalogue.
seed-locked-iteration
Use the seed parameter to isolate what a prompt change actually did, instead of rolling new dice every generation. Use whenever the user is iterating on an APImage generation, says output is inconsistent or unpredictable, mentions seeds or reproducibility, or can't tell whether an edit improved anything.
shopify-product-images
Build product imagery for a Shopify store — variant images, collection consistency, theme aspect ratios and performance. Use whenever the user runs a Shopify store, is preparing product images for their own site, or asks about variant images or collection page consistency.
social-commerce-product-images
Produce product imagery for social feeds and paid social, where native and scroll-stopping beats catalogue-clean. Use whenever the user needs product images for Instagram, Facebook, Pinterest, TikTok Shop, paid social ads, or social commerce placements.
talking-head-scripting
Write scripts for spokesperson and avatar video that fit the duration, sound like speech, and survive lip-sync generation. Use whenever the user is writing a script for a talking-head video, an avatar presenter, a voiceover, or has a script that runs too long or sounds written.
text-to-video-prompting
Write text-to-video prompts that produce the motion, camera and subject behaviour you asked for. Use whenever the user is generating video from a text prompt with APImage, gets drifting or morphing or static output, or asks how to prompt for camera moves and motion.
tiktok-video-generation
Generate video for TikTok specifically — native formatting, hook conventions, sound-off viewing, and what the platform's own rules require of AI content. Use whenever the user wants TikTok videos, TikTok Shop content, TikTok ads, or asks how to make generated video work on TikTok.
trend-format-replication
Adapt a trending short-form format to your product without copying the content or infringing rights. Use whenever the user wants to use a trend, replicate a viral format, mentions trending audio or a viral template, or asks how to make generated video feel current.
ugc-ad-video-generation
Generate UGC-style ad video from a product photo — the core APImage workflow — with the structure, look and disclosure that make it work. Use whenever the user wants UGC ads, creator-style video, authentic-looking product ads, or asks how to turn a product photo into an ad video.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.