Agent Skills · Design & Media
Curated skills for Claude, Cursor and other AI agents. Copy one prompt to install.
Turn finalized columns or spoken notes into serialized corporate podcasts with voice cloning, directory voices, pronunciation sheets, episode ledgers, and block-level re-synthesis.
Generate zoned garden plans with climate-appropriate plants, hardscape, drainage, lighting, and phased budget estimates.
Orchestrates AI short-drama clips, subtitles, and voiceovers into a local, offline FFmpeg editing project with transitions, subtitle burn-in, audio mixing, BGM, self-checks, presets, and publish handoff.
Analyzes video or stills across 13 dimensions and generates multi-platform AI prompts, with copyright checks and image-to-video guidance.
Upload cover images to obtain preflight checks and candidate versions, optimizing visual hooks, focus hierarchy, and mobile readability.
Generate 8 to 20 tagged game UI voice lines from provided interface copy, covering buttons, wins, and losses.
A general 3D space, movement, and camera scheduling system for film, animation, game CG, and AI visuals that outputs staging, paths, and prompts.
Guides long-form video to vertical clips, batch edits, and AI highlights with tool choices, parameters, SOPs, and ffmpeg scripting guidance, without rendering video files.
Use natural language to drive local Photoshop CS6 and run ExtendScript for posters, cutouts, watermarks, and batch resizing; if unavailable, output a .jsx script and manual steps.
Recognizes Japanese video audio, translates it into Chinese subtitles, and burns them in via Whisper, LLM/MyMemory, and ffmpeg.
An AI image batch generation tool for TikTok cross-border e-commerce, supporting ERP product lists, multilingual site adaptation, and product hero, scene, and detail images.
Use the dLazy Vidu audio clone API to clone a voice and generate new text-to-speech audio from prompts.