Agent Skills · Design & Media
Curated skills for Claude, Cursor and other AI agents. Copy one prompt to install.
Generate or edit high-quality images via the dLazy CLI with Nano Banana 2.0, supporting text-to-image and image-to-image.
Recognizes geometry and dimensions from engineering images, generates annotated DXF files with previews, and outputs a clean DXF containing only outline, inner, and hole layers.
Supports single or batch text-to-image generation and text-to-video generation with AI audio via the Zhipu BigModel API.
Use Tencent Hunyuan to generate 3D models from text or images, with PBR materials, custom face counts, and multiple output formats.
Generate short videos, posters, infographics, and product images via the TianMei AI API, with WeChat login, credit top-ups, and digital human authentication.
Download audio from any URL and transcribe it to plain text with Faster-Whisper, with model, language, and quantization options.
Build a design philosophy first, then render a minimalist 3:4 SVG poster.
Use Tencent Cloud Hunyuan to generate images from text, with reference images, custom resolution, fixed seed, and prompt revision.
Generate complete, playable, and downloadable songs from a theme or lyrics, then return project links and two candidates synced to Haimian Music for mini-program and web management.
Generates bilingual, structured text-to-image prompts with subject, style, lighting, composition, and technical specs for GPT-4o, Midjourney, DALL-E, and domestic platforms.
A PM-led product design workflow that moves from requirements and metrics to single-HTML prototypes, iterative polish, dashboards, and demo-ready deliverables.
Generate videos using Volcengine's SD1.5pro API, supporting text prompts and image references.