skills.rest
inclusionAI
inclusionAIOFFICIAL
Explore all skills by this author →
WebsiteGitHub

ad_image_create

Generate ad-ready product images and collages with computed sub-image sizes and social taglines.

1.2k|126|Updated Mar 14, 2025
Search Tags:#image generation#product marketing#collage#tagline#ad image#Social Media#Scene Design
One-click install
npx skills add https://github.com/inclusionAI/AWorld --skill ad-image-create
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ad_image_create
Source: https://github.com/inclusionAI/AWorld/tree/main/aworld-skills/ad_image_create_skill
Command: npx skills add https://github.com/inclusionAI/AWorld --skill ad-image-create

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires Pillow.

What problem does it solve?

Creating advertising images from product photos requires coordinating scene design, image generation parameters, collage layout math, and social copy, which is tedious to do manually and error-prone when sub-image sizes must match a target output ratio.

Core Features & Use Cases

  • Ratio-driven size back-solving: Computes sub-image dimensions from the final target ratio and collage layout (2x2, 1x3, 1x2) so generated images stitch together without manual cropping.
  • Structured image generation: Calls image_generator with a strict request contract (content, image_url, size, output_dir, optional reference_images) and enforces explicit actor-count language to prevent unwanted extra subjects.
  • Product-grounded scene design: Activates media_comprehension to extract product style, tone, and audience before designing scenes, and pairs each deliverable with a short 小红书/抖音-style tagline.
  • Use Case: Given one product photo and a 1920x1080 target, generate four 960x540 scene images, stitch them into a 2x2 collage, and attach one social tagline for the final ad.

Quick Start

Create a 16:9 collage ad from my product photo with four lifestyle scenes and a short tagline for 小红书.

Frequently Asked Questions about ad_image_create

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a collage ad image from a product photo?▼

Decide the final ratio and layout first, then compute sub-image sizes (for example 960x540 for a 2x2 grid at 1920x1080). Generate each sub-image with image_generator using those exact sizes, then stitch them onto a canvas with PIL.

How do I control the number of actors in generated ad images?▼

Use explicit count language in the prompt, such as "only one" or "at most two" in English, or 只有一只/最多两只 in Chinese. Without explicit counts, the model often generates extra actors in the scene.

Can I use a video as a reference for the supporting actor?▼

Yes. Capture frames from the user-supplied video, run media_comprehension on candidate frames to confirm the desired actor appearance, and pass the chosen frame as reference_images to image_generator.

What sub-image size should I use for a 1080x1080 collage?▼

For a 2x2 layout use 540x540 per sub-image, and for a 1x3 layout use 360x1080. The size string passed to image_generator must match the computed sub-image dimensions exactly.

Why does my stitched collage not match the target size?▼

Generator output can drift slightly in dimensions, for example a small height offset. Validate the final stitched output against the requested ratio and crop after stitching if needed.

Related Skills

Explore More Marketing & Sales Skills→
SamurAIGPTSamurAIGPT
4.2k

muapi-freeze-effect-video

Generates a cinematic freeze-effect video from a reference photo using the muapi CLI.

Official
Basic
itwangeritwanger
17.5k

video-cover-image

Generate matched 3:4, 16:9, and 4:3 video cover images from technical scripts.

Community
Advanced
Largo2z9Largo2z9
1

qc-creative

Validate rendered creative assets for product fidelity and compliance risks.

Community
Advanced
runcomfy-comruncomfy-com
12

flux-kontext

Edit single images with Flux 1 Kontext Pro via RunComfy CLI.

Official
Intermediate