Pack

Seedance 2.0 Skill OS

Emily2040/seedance-2.0

A large skill set for writing Seedance 2.0 video prompts: a routing root skill plus 28 sub-skills for camera, audio, character locks, continuation and troubleshooting.

View on GitHub
01Author
Emily2040
03Stars
7,501 · as of Oct 3, 2026
04Last updated
Sep 26, 2026
05License
MIT
06Works with
Claude Code · Codex · Cursor · Gemini CLI · OpenClaw
07Models
Seedance 2.0
08Content language
English and Chinese
09Domains
AI video · Filmmaking · Short drama
10Stages
Prompting · Storyboard
SC 01Overview

What it does

Given an idea, a draft prompt, reference files or a failed take, the root skill checks the target platform, picks a mode (text, image, video reference, first/last frame, edit or extend) and routes to sub-skills for camera, lighting, motion, audio and lip-sync, character locks, visual style, IP-safe rewrites and moderation-block diagnosis. Longer stories are split into connected clips, and each continuation prompt is written from the real ending of the footage you accepted. The repository also ships genre recipes (including a short-drama pattern), vocabulary packs in six languages and example banks in Chinese, Japanese and Korean. It prepares prompts and plans rather than rendering video, so generation and its cost stay with your video provider.

Included skills

  • seedance-20
  • seedance-prompt
  • seedance-prompt-short
  • seedance-interview
  • seedance-interview-short
  • seedance-sequence
  • seedance-continuation
  • seedance-camera
  • seedance-lighting
  • seedance-motion
  • seedance-vfx
  • seedance-audio
  • seedance-characters
  • seedance-style
  • seedance-recipes
  • seedance-pipeline
  • seedance-troubleshoot
  • seedance-copyright
  • seedance-filter
  • seedance-antislop
  • seedance-vocab-en
  • seedance-vocab-zh
  • seedance-vocab-ja
  • seedance-vocab-ko
  • seedance-vocab-es
  • seedance-vocab-ru
  • seedance-examples-zh
  • seedance-examples-ja
  • seedance-examples-ko
SC 02Install

How to install it

  • 01From the README

    The install command quoted from the repository README.

    git clone https://github.com/Emily2040/seedance-2.0.git cd seedance-2.0 python scripts/install_codex_skill.py --client codex --scope user
  • 02skills CLI, for Claude Code, Codex and others

    The open skills CLI from Vercel Labs installs a skill from a GitHub repo into the folder each agent expects. It asks which skills and which agents you want, and it installs for the current project unless you add the -g flag, which installs for your user account. Add --list to see what a repo contains before you install anything.

    npx skills add Emily2040/seedance-2.0
  • 03Claude Code, skills folder

    Copy the skill's folder, the one that holds SKILL.md, into ~/.claude/skills/ to use it in every project on your machine, or into .claude/skills/ in a repository to use it there and share it by committing it. Claude Code picks up changes to these folders during a session. If the folder did not exist when the session started, run /reload-skills.

  • 04Claude Code, plugin marketplace

    Some repos publish a plugin marketplace, a .claude-plugin/marketplace.json file. Add it in Claude Code, then install a plugin from it with /plugin install plugin-name@marketplace-name. Plugins can bundle hooks and MCP servers along with skills, so read the details panel before you confirm.

    /plugin marketplace add Emily2040/seedance-2.0
  • 05Codex, skills folder or $skill-installer

    Codex reads skills from .agents/skills in your working folder and its parent folders up to the repository root, and from $HOME/.agents/skills for every repository. Copy the skill's folder into one of them, or run $skill-installer inside Codex and ask it to install from the repo. Codex detects new skills automatically and needs a restart only if one does not appear.

The first command is quoted from the author's README. The others are the standard ways to install a skill from a GitHub repository. They are templates and may not apply here, for example if the repo has no plugin marketplace, so follow the README first.

SC 03Guide

How to get the most from it

Good for

  • Writing a single-clip Seedance 2.0 prompt with one visible action, a clear ending, a fixed camera choice and an explicit sound plan.
  • Planning connected multi-clip stories, where each next prompt is built from the footage you accepted rather than from the original plan.
  • Assigning one primary role to each image, video or audio reference and keeping reference tags unchanged across prompts.
  • Diagnosing a take that missed or was blocked by moderation, and rewriting wording that touches protected characters, brands or real people.

Watch out for

  • Every number in it (durations, reference limits, model IDs) is a Seedance 2.0 value; the README treats Seedance 2.5 as a separate line, so check which one your provider runs.
  • It prepares prompts and plans; it does not render video. You generate on your own provider account and pay for it there, and installing needs a local clone plus Python for the bundled installer.
  • The root skill loads many reference files depending on the request, and the README says live scoring of rendered results is still pending, so treat it as a method rather than a quality guarantee.

Typical workflow

01Step
Clone the repository and run its installer for your client (Codex user scope, Claude Code, or another skills folder), then restart the client and select seedance-20.
02Step
Tell the agent what happens in the clip and what must stay fixed, for example one static shot and no music, and ask for the prompt only.
03Step
If you have image, video or audio references, provide the files and say what each one should control; the skill gives every asset one primary role.
04Step
For a longer story, describe the full idea and its ending; the skill splits it into clips and a project state, then writes Clip 01.
05Step
Generate the clip on your provider, send back the result or its final frame, and let the skill record the real ending before it writes the next clip.
06Step
If a take misses, give the failed requirement and how many retries you have left; the skill diagnoses it and proposes a retake that changes one variable.
SC 04Fit

Where it fits in a film pipeline

The production stages this skill covers, and where Pencil0 picks up.

SC 05Related

Related skills

  • Official

    OpenDesign video-shortform

    nexu-io/open-design

    A skill bundled with the OpenDesign app that plans one short AI video shot, writes the prompt the way the chosen model prefers and renders it through the app's media command.

    • Generation
    • Prompting
    99.2kGitHub starsUpdated Oct 2026
  • OfficialPack

    Toonflow Skills

    HBAI-Ltd/Toonflow-app

    Skills that come with the open-source Toonflow canvas app: a production workflow with Seedance 2.0/2.5 prompt rules, a canvas manual, a text-storyboard skill and an MCP skill for outside agents.

    • Pipeline
    • Prompting
    • Storyboard
    16.3kGitHub starsUpdated Oct 2026
  • OfficialPack

    Huobao Drama Skills

    chatfire-AI/huobao-drama

    Seven skills from the Huobao Drama app, in Chinese with English variants: novel-to-script rewriting, asset extraction, storyboard segments, and character, scene, prop and video prompts.

    • Prompting
    • Storyboard
    • Screenwriting
    15.6kGitHub starsUpdated Oct 2026
  • OfficialPack

    waoowaoo creative skills

    waooAI/waoowaoo

    Six Chinese-language skill files from the waoowaoo AI film workspace, covering script writing, creative direction, asset design, video-segment prompts and music cue planning.

    • Prompting
    • Screenwriting
    • Storyboard
    14.3kGitHub starsUpdated Sep 2026
  • OfficialPack

    MiniMax H3 skills

    MiniMax-AI/MiniMax-H3

    MiniMax's official skills for its H3 video model: one portable prompt-writing skill plus eight style-specific video workflows that need the MiniMax Hub agent.

    • Prompting
    • Pipeline
    • Generation
    9.5kGitHub starsUpdated Aug 2026
  • Official

    Gemini Omni Flash API

    google-gemini/gemini-skills

    Google's skill for generating, extending and editing video with Gemini Omni 1.1 Flash through the google-genai SDK, with scripts for upload, ffmpeg prep and batch runs.

    • Generation
    • Prompting
    4.2kGitHub starsUpdated Sep 2026

This is a third-party skill. It is not maintained or audited by Pencil0, and skills can run scripts on your machine, so read the SKILL.md and any scripts before installing. Stars, license and dates are a GitHub snapshot and may be out of date.