AI & Machine Learningttfake92-lab/skills
yyl-video-prompt
Yuyile's flagship AI video prompt Skill. First select a tier based on the script style, then act like a director to turn scripts, stories, voice-over scripts, shot lists, or a single idea into segmented video prompts ready for direct generation. The short-video tier (Douyin/Xiaohongshu pacing: one continuous shot, multi-shot, street footage, character micro-stories, vertical-format short dramas; benchmarks are *Danmei* and *Cyber Atom*) and the cinematic premium tier (slow-paced narrative, acting, dialogue, two-hander scenes, AU micro-expressions, eight-dimension formula) each have their own structures and examples. Every segment includes the generation mode and parameters, reference-image bindings, characters, props, and setting, shot timeline, sound, and hard constraints. Compatible with Seedance 2.0/2.5, Jimeng, dreamina, LibTV, and TT AI. Use when the user wants to write video prompts, AI video prompts, Seedance prompts, Jimeng prompts, convert scripts to video, break down storyboards, write storyboard prompts, create fast-cut videos, write short-video prompts, create a storyboard or shot list, use multi-image references, work with first-and-last frames, write character acting prompts, explore AI acting, emotional progression, director-style prompts, or provide a script, story, or idea to turn into video. If the current project has a dedicated workflow Skill (Vlog, emotional hooks, news commentary, etc.), prioritize the dedicated one.