Seedance 专业级导演全流程：收稿、鞭打、讲戏、资产、镜头表、编译九段。即梦 Seedance 2.5 / 2.0 提示词工程。
Plugin seedance-daoyan from marketplace seedance-daoyan in aaronbedolla560-cell/Seedance-Director-4.0. Commands read from the repo's manifest — 26 days ago.
bianda鞭打。要把这部片子的片种、意图、风格、范围钉成下游不用猜的蓝图时用。改已有蓝图、审结构、找弱场也用。不改剧本、不拆条、不写提示词。
bianju编剧。双确认过后开集，把定稿剧本翻成讲戏本时用。逐场讲戏、拆条、预算时长、提出资产清单；首集把风格草案定稿。不改剧本、不写提示词、不定镜头数。
bianyi编译。把已确认的镜头表编译成目标平台动态提示词时使用。本站是编译器，不改故事、不增删镜头。默认 Seedance 2.5 九段；2.0 改块序不改一条一意。
fangong返工。用户带着生成翻车的视频条或图片回来报修时使用。诊断症状、五出口路由、就地更新产物与版本注记。生成前的审核 FAIL 不进本站。
gongyi工艺。写镜头表、编译动态提示词、或返工修运镜/空间/表演/光/声时加载。五维写法合在一个工艺组，按需翻章，不单独走流程，不独立成片。
jiaojihua交际花。拷问之后把剧本或小说改成可拍文字时使用。结构、可视化、可实现性三类改造一次做完。剧本改动的版本增量也走本站。
jinghua自净化。用户带反馈回来（翻车、吐槽、新需求、好案例）要把经验沉淀进技能库时使用。把单次教训翻译成可复用规则，改的是库不是本片。与 fangong 的分工：fangong 修片子，jinghua 修系统。
jingtoubiao镜头表。资产确认后为这部片子落全集镜头表，或用户指镜号改、说这场重讲时使用。钉死每场几条几镜、轴线、控制等级。
kongzhitu控制图。镜头表逐镜标注的控制等级需要图，或用户提到关键帧、首帧、宫格、站位图、白模、深度图、多人站位压不住、构图漂移、空间动线控制时使用。产出 05-control-prompts/epNN-sXX-control.md。本站不生图、不写视频成品九段。
lianxu连续。编译段走完一场要更新状态台账、集末要登记跨集落幅、fangong 修复涉及状态变化，或成片审需要连续性批检方法时使用。维护 07-continuity/ledger.md。台账只记事实，不改成品。
qiangshou枪手。这部片子只有梗概、想法或大纲，没有可拍剧本时使用。按蓝图代写场，支持全本写与逐场共创。不发明要拍什么，不宣告定稿。
shenhe审核。拆解、资产包、全集镜头表、全集提示词四类产物落盘，或用户点名质检时使用。永远派 fresh zhaocha，只审不改。产出 .review/ 报告与回流路由。
shougao用户带着剧本、小说、长文本或任何故事文字进入这部片子时使用。只判定类型并原样存档，不改一个字。想法也存，再转 bianda。
yuyan预演。用户想在花钱生成之前先看画面大概是什么样时使用。把已写死的镜头表翻成三维场景 JSON，交 previz.py 渲成导演台草图。可选、不设门禁、不进交付。
zichan资产。拆解确认后一次性出角色、场景、道具三包时使用。表审后新增或状态变体走增量。图像提示词，纯正向。
ziqu自取。用户想把一段流程目标交给 /goal 自驱执行时使用。编译一条带预授权范围、四个审核点与逐站确认处置、可自证完成条件的 /goal 指令，交用户复制发送。本站不替用户触发。
Build an immersive scroll-scrubbed landing page where a pre-rendered camera flies from outside each scene into its interior, then flows on to the next with no cuts. Interviews the user for topic, story beats, and brand kit; generates cohesive scenes + seamless camera clips with Higgsfield; wires a portable, framework-agnostic scrub engine.
1 skill
Make AI UGC videos end to end through treg. Pull the trending TikTok and Instagram videos in a vertical, extract the hook patterns, create a character with the same vibe as a presenter the user picks, generate 5-10 talking-head hook clips on Seedance 2.5 (less-restriction route), add a voiced demo clip, burn captions, and report the bill per clip. Use when the user asks for UGC ads, creator-style product videos, TikTok/Reels hooks, or an AI presenter for their product.
1 skill
Generate AI images, videos, music and speech from Claude Code through the PopNow API: Nano Banana, GPT Image, Seedream, FLUX, Kling, Seedance, MiniMax and Suno behind one key, plus workflow skills for launch films, explainer videos, social covers, landing pages and brand kits. Every paid run is quoted in USD and waits for an explicit yes. Needs a PopNow API key with a prepaid balance.
32 skills
Two stages: first a single-page sheet of numbered panels from an image model, then that sheet fed to Seedance as the reference. The sheet owns order, framing and timing, and the video prompt only has to connect the panels.
1 skill
Agent skills for Claude Code, Codex, Cursor and Gemini CLI that make a multi-shot AI short film from an idea (story, shot sheet, takes, voice, music, edit) and run Seedance, Kling and Wan video models, with the price shown before every paid job.
2 skills
AI video generation — describe what you want, Pexo picks the best model across 10+ engines (Seedance, Kling, Veo, Sora) and returns a finished, multi-shot video with music, subtitles, and transitions. Includes the Pexo agent plus image, audio, director, and model-prompting skills.
20 skills
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.