Agent Skills · Design & Media
Curated skills for Claude, Cursor and other AI agents. Copy one prompt to install.
Generate consistent 3D asset reference renders from text or photos, including clay, multi-view, PBR, topology, UV, exploded, and keyframe outputs, without direct 3D files.
Wraps the Agnes AI platform API for text-to-image, image editing, and video generation with async task polling.
Extract character profiles from Chinese or English scripts and generate white-background nine-grid, facial close-up, and multi-emotion prompts for consistent AI image generation.
Provides ID photo dimensions, background, shooting, and layout requirements, plus tool recommendations; it does not edit images and suggests studios for important documents.
Selects the best dLazy CLI TTS, music, sound-effect, or voice-cloning model from a prompt and runs generation through the dLazy API.
Call the dLazy Jimeng i2v-first API to generate a dynamic video from a first-frame image and prompt.
Asynchronously generates personalized wedding invitation images from wedding photos and event details, with style analysis, layout copy, and size options.
Guides engineers in writing Seedance 2.0 multimodal video prompts with @ references, camera motion, timing, audio, and style for ads, short dramas, and explainers.
Generate images via the LiblibAI ComfyUI App API, supporting public URLs, local R2 uploads, and Feishu media returns.
Use a local CLI to generate previewable and downloadable Suno songs from themes, lyrics, styles, or reference audio, then manage them in the Haimian Music account.
Generate GPT images via the OpenClaw API with text or reference images, quality tiers, credit estimates, task polling, and multi-image sets.
Uses dLazy Kling SFX to generate matching scene sound effects from text descriptions or video frames.