Agent Skills · Design & Media
Curated skills for Claude, Cursor and other AI agents. Copy one prompt to install.
After image safety review, asynchronously generate a 3x3 hairstyle design preview from a portrait and split it into nine individual images.
Use the dLazy API to convert images into dynamic dance videos with Seedance 1.5 Pro.
A local audio/video transcript organizer that accepts media, links, subtitles, or raw drafts and produces a cleaned share-ready Chinese original transcript.
A design spec for ice-blue frosted dashboards, covering backgrounds, glassmorphism, chart choices, grid layout, and interaction details.
Generate SVG vector graphics from text with Recraft V3 via the dLazy API.
Use Hua & Hua brand-strategy methods to define problems, match frameworks, generate prioritized actions, and support execution with templates and measurable outputs.
An Agent-facing image editing API that accepts Base64 or URL reference images, uses async task polling and WeChat Pay retry, and returns generated image URLs.
A skill providing a structured frontend design workflow, covering layout, theme, animation, and implementation for building modern UI components and pages.
Generate style designs from 100 six-slot schemes through subject swaps, element remixes, template derivatives, style extensions, and domain mappings.
Generate serial audiobook chapters from finalized webnovel text with a fixed narrative voice and pronunciation glossary.
A design-first prompt workflow for GPT Image 2.0 that analyzes visual strategy and outputs copy-ready positive prompts for image generation, reference images, inpainting, and prompt reversal.
Create digital human videos from portraits and audio with Jimeng OmniHuman 1.5.