Agent Skills · Design & Media
Curated skills for Claude, Cursor and other AI agents. Copy one prompt to install.
Let agents run image-to-image generation with Jimeng CLI, using local reference images, explicit resolution and size, and polling task completion.
Handles text-to-image and image-to-image requests with selectable aspect ratios and 1K/2K/4K output sizes, returning structured results.
AI-assisted image-and-text logo design with infringement pre-checks, free-font recommendations, design specs, and trademark delivery guidance.
Use dLazy CLI with Wan 2.6 Flash to quickly turn reference images into dynamic short videos, with async task polling support.
A Jimeng text-to-image client powered by the dLazy API for converting prompts into high-quality images.
Use the dLazy hosted API to call Google Veo 3.1 for cinematic text-to-video and image-to-video generation, with async task and local file upload support.
Labeled original instrumental playlist for livestream and store slots.
An illustration skill for engineers that generates white-background, thin-line doodles with black-bean characters and sparse orange, red, or blue accents for retros, travel notes, process explainers, and light posters.
A per-call paid image-generation API for agents, supporting text-to-image, reference-guided editing, multi-reference consistency, in-image text, and async polling.
Extracts Bilibili video speech, prefers CC subtitles when available, otherwise transcribes with Whisper, and outputs a structured transcript with timestamps, sections, and key-point tables.
Calls ByteDance's Seedance 2.0 through dLazy to generate videos from images, video, audio, first/last frames, or text.
Generates role-labeled, line-ordered voice packs for script-kill or game scripts with fixed character voices and per-block synthesis.