Agent Skills · Design & Media
Curated skills for Claude, Cursor and other AI agents. Copy one prompt to install.
Selects the best dLazy CLI video model for text-to-video or image-to-video generation.
Generate three differentiated cinematic fashion editorial shoots and plans from one portrait photo, with face consistency and compliance review via ShowAPI async tasks.
A thin client for the dLazy Happy Horse 1.0 video model, covering text-to-video, image-to-video, reference-to-video, and video editing via the dLazy API.
Wraps MiMo V2.5 TTS scripts for preset voices, voice design, voice cloning, natural-language control, audio tags, dialects, and singing.
Resolve Bilibili video links, download audio and video, generate timed subtitles with Whisper, analyze keyframes, and produce a structured summary.
A design-methodology engine for interior and spatial projects, generating culture-driven concepts and organizing deliverables into Z0-Z3 structured documentation.
Generates an algorithmic art philosophy from user input, then builds a p5.js HTML artifact with seeded, parameter-controlled generative art.
Wraps the Chilltion AI video API with shell scripts for session creation, SSE progress, and video URL retrieval.
Rebuild an authorized template's editing structure with your own footage, captions, and narration timing to produce an editable Jianying draft.
A design skill for generating SOLIDWORKS parts, assemblies, drawings, BOMs, or concept STEP/GLB models with four views from natural-language requests.
Compress PNG files with Python/Pillow using lossless or 256-color lossy modes, supporting single-file or batch directory processing.
Generate a three-minute book explainer video from a book title and author, covering copy, scenes, images, TTS, subtitles, and composition.