Agent Skills · Design & Media
Curated skills for Claude, Cursor and other AI agents. Copy one prompt to install.
Turn topics or materials into renderable PlantUML mind maps with hierarchy, direction, multiline nodes, colors, and common patterns.
Extracts research structure from documents or text and generates academic technical roadmaps using SVG, Seedream, or Gemini.
Convert photos into cute chibi or doodle illustrations and frame them on an Ardot canvas for preview and sharing.
Converts plot text to Seedance 2.0 reference video prompts.
Explains abstract professional concepts through analogies, text-based mind maps, layered explanations, and memory tips, while noting limits over precise definitions and real charts.
Generate a 3x3 outfit grid from a portrait and style prompt, then split it into nine independent images.
Derives a design philosophy from user input and generates visually driven canvas artwork as a Markdown brief plus PDF or PNG output.
A Zhipu AI API-powered tool for text-to-image, image-to-video, watermark removal, and local image upload workflows.
An agent-ready workflow that converts scripts into HTML, CSS, and GSAP video compositions and renders deterministic MP4 output using Headless Chrome and FFmpeg.
Call a local music-generation API via CLI to create songs, lyric-based tracks, BGM, and adapted versions, returning playable and downloadable links.
Turns daytime office briefs into a tagged, low-stimulation instrumental playlist of 8 to 15 background tracks.
A Plane Wrap wrapper for GitHub automation.