Image
hyperframes-studio keeps HyperFrames timelines editable: every scene its own sub-composition, one caption track, 90%/80% safe boxes. 370,619 installs.
HyperFrames’ general-video skill routes by project state, then fans out scene packets only past six scenes: 9 min inline vs 21 min dispatched. 594K installs.
ian-xiaohei-illustrations turns Chinese articles into 16:9 hand-drawn Xiaohei images, guided by a shot-list workflow and strict white-background style DNA.
Guizang Social Card Skill turns articles into Xiaohongshu carousels, Live Photo motion cards, and WeChat 21:9+1:1 cover pairs with 28 layouts and 10 themes.
Guizang PPT Skill generates single-file HTML decks with magazine or Swiss design, presenter mode, rehearsal tools, and social covers for AI agents.
Design mobile apps with Sleek’s official agent skill — create projects, direct AI screen design, and implement results in React Native or SwiftUI.
HeyGen’s official agent skills — heygen-avatar, heygen-video, heygen-translate turn a photo into a reusable avatar that delivers videos in 175+ languages.
Pika Labs turns text and images into short AI videos — Pikaffects, Pikascenes, and an MCP that gives agents 42 media tools for ads, podcasts, and launch clips.
Generate production-ready UI mockups and design assets directly from your AI coding agent using Google Gemini’s multimodal vision and generation capabilities.
Claude Design is a portable Claude skill that turns Claude into an expert designer for HTML artifacts — landing pages, slide decks, interactive prototypes, animations, and more.
Kling 3.0 exposes six video generation endpoints on RunComfy — Standard, Pro, and 4K in text-to-video and image-to-video modes, all from a single skill.
GPT Image 2 brings OpenAI’s strongest text-to-image model to your terminal through RunComfy — no ChatGPT tab, no API key management, one CLI command.