Agent Skills · Design & Media
Curated skills for Claude, Cursor and other AI agents. Copy one prompt to install.
Detects available compression tools, selects strategies by image type, and processes folders in parallel with quality, preview, and output options.
Generates upload-ready e-commerce scene main images from product references, preserving appearance and brand marks, and returns result image URLs via async tasks.
Generates 4–8 consistent branch lobby window boards from supplied window names and key points without inventing rates, hours, rankings, or seals.
A UI/UX skill for Chinese-style mobile games that outputs Figma design tokens, Godot node trees, and shader parameters.
Works with the JMCAI Comfypet desktop app to select workflows, submit alias-schema runs, and poll status and local outputs.
Use aimv to remix demos, reference audio, or songs, then download results.
Use the dLazy API to generate diverse design-style images with Recraft v3 text-to-image and image-to-image workflows.
A Volcengine Seedance 1.0 Pro text-to-video tool that generates 4-12s 1080p MP4 videos and downloads them after polling.
Wraps the redfox.hk Seedream 5.0 Lite image API to generate images via one command, supporting text-to-image, image editing, sequential sets, and HD output.
Suggests natural photo poses by scenario, group size, and vibe, with body positioning, composition, and lighting tips.
Uses ffmpeg to extract random clips from filtered videos, merge them, add a transparent logo and random background music, and batch-render final outputs.
AI image upscaling with Real-ESRGAN and LANCZOS, supporting PSD/PNG/JPG, batch output, and DPI metadata.