Agent skill

Script To Shootable Storyboard

by zyz254009-crypto in zyz254009-crypto/script-to-shootable-storyboard

将剧本、故事梗概、对白稿、已有分镜或视频生成提示转换、检查并优化为可拍、可剪、适合 Seedance、真人或混合制作的原子分镜。用于剧本转分镜、分镜可拍化、后续场景开发、反馈驱动迭代、不连贯诊断与修复、心理描写视觉化、单镜主导景别拆分、短剧钩子与节奏、台词和声音时长、人物道具连续性、Seedance Generation Task、生成风险与失败降级;默认忠实转换,未获授权不得改写剧情。

MITAuto-check passedMedia & Creative

Install Script To Shootable Storyboard

skills CLI
$ npx skills add zyz254009-crypto/script-to-shootable-storyboard --skill script-to-shootable-storyboard -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install zyz254009-crypto/script-to-shootable-storyboard script-to-shootable-storyboard --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
script-to-shootable-storyboard
GitHub stars
117
Token cost
~1.9k tokens
SKILL.md length
310 words
Files
95 (incl. scripts, references, assets)
Skills in repo
1
Repo updated
First seen
Licence
MIT

At a glance

将剧本、故事梗概、对白稿、已有分镜或视频生成提示转换、检查并优化为可拍、可剪、适合 Seedance、真人或混合制作的原子分镜。用于剧本转分镜、分镜可拍化、后续场景开发、反馈驱动迭代、不连贯诊断与修复、心理描写视觉化、单镜主导景别拆分、短剧钩子与节奏、台词和声音时长、人物道具连续性、Seedance Generation Task、生成风险与失败降级;默认忠实转换,未获授权不得改写剧情。

  • Works in 5 steps: 识别输入类型:script、scene_draft、synopsis、dialog… → 建立 JobSpec;从 project-input-template.yaml… → 锁定修改权限 → …
  • Tasks that involve AI video generation
  • SKILL.md covers 先锁定任务, 最小读取路径, 执行主流程 and 强制可拍规则, plus 9 more sections
  • Calls node

What it does

Script To Shootable Storyboard is an agent skill from zyz254009-crypto/script-to-shootable-storyboard. 将剧本、故事梗概、对白稿、已有分镜或视频生成提示转换、检查并优化为可拍、可剪、适合 Seedance、真人或混合制作的原子分镜。用于剧本转分镜、分镜可拍化、后续场景开发、反馈驱动迭代、不连贯诊断与修复、心理描写视觉化、单镜主导景别拆分、短剧钩子与节奏、台词和声音时长、人物道具连续性、Seedance Generation Task、生成风险与失败降级;默认忠实转换,未获授权不得改写剧情。

Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 97 other files, including scripts, reference files and assets (for example `PROJECT_OVERVIEW.md`, `README.md` and `agents/openai.yaml`).

It sits in Media & Creative, covering AI video generation and Comics and storyboards. It works with Seedance and Node.js. The repository describes itself as: 使用 Codex 开发的剧本转可拍分镜 Skill,在 Codex 中运行效果最佳,支持 Seedance、真人/混合制作与二次开发。 The licence is MIT.

When your agent uses it

  • Tasks that involve AI video generation
  • Tasks that involve Comics and storyboards

Example prompts

  • “/script-to-shootable-storyboard”

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. 识别输入类型:script、scene_draft、synopsis、dialogue_draft、existing_storyboard、seedance_prompt 或混合输入。
  2. 建立 JobSpec;从 project-input-template.yaml 起步。
  3. 锁定修改权限
  4. 选择 live、seedance、hybrid 或逐镜决策。
  5. 冻结原稿、版本、来源范围、锁定事实、假设和未知项。

What it can do on your machine

Read from SKILL.md and the folder at commit e1ee698. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 1 file in scripts/, which the agent can run.

    Shell commands in SKILL.md call:

    • node

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Script To Shootable Storyboard loads about 1.9k tokens when it runs, and up to ~180k if it reads all its reference files. Until then it costs about 57 tokens; SKILL.md has 310 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~57
When it runs · the whole SKILL.md, loaded when a task matches
~1.9k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~180k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from zyz254009-crypto/script-to-shootable-storyboard at commit e1ee698, republished under its MIT licence (© zyz254009-crypto). 310 words, ~1,945 tokens.

Download SKILL.mdSave it as .claude/skills/script-to-shootable-storyboard/SKILL.md (or your agent's skills folder). This skill also uses 94 other files; get the full folder from GitHub.
name
script-to-shootable-storyboard
description
将剧本、故事梗概、对白稿、已有分镜或视频生成提示转换、检查并优化为可拍、可剪、适合 Seedance、真人或混合制作的原子分镜。用于剧本转分镜、分镜可拍化、后续场景开发、反馈驱动迭代、不连贯诊断与修复、心理描写视觉化、单镜主导景别拆分、短剧钩子与节奏、台词和声音时长、人物道具连续性、Seedance Generation Task、生成风险与失败降级;默认忠实转换,未获授权不得改写剧情。

剧本转 Seedance 可拍分镜

把输入编译成摄像机、演员、声音、剪辑和 Seedance 能执行的镜头方案。优先保证剧情保真、画面可见、动作可执行、镜头原子化、声音可定位、连续性可追踪和失败可降级。

先锁定任务

  1. 识别输入类型:script、scene_draft、synopsis、dialogue_draft、existing_storyboard、seedance_prompt 或混合输入。
  2. 建立 JobSpec;从 project-input-template.yaml 起步。
  3. 锁定修改权限:
    • faithful:默认;只拆分、画面化和最小必要补全。
    • visual:可增强动作、构图、道具、声音和环境反应,但不新增剧情事实。
    • pacing:可拆合镜头、压缩和局部调序,但不改变核心因果。
    • story:仅在用户明确授权和提供授权范围后改写剧情。
  4. 选择 live、seedance、hybrid 或逐镜决策。
  5. 冻结原稿、版本、来源范围、锁定事实、假设和未知项。

读取 workflow-modes.md 处理输入分流、权限和已有分镜优化。不得把“优化分镜”自动解释为“改写剧情”。

最小读取路径

执行主流程

按顺序执行,不要让下游补丁掩盖上游错误:

  1. 冻结来源并建立 Source Manifest、Fact Ledger 和字段级 provenance。
  2. 解析 Scene、人物、地点、时间、道具、对白和声音。
  3. 建立 Story Bible,锁定人物身份、关系、外观锚点、知情范围和世界规则。
  4. 按行动、信息、权力关系或局势变化拆 Beat;不要按句号或台词数量拆。
  5. 从原剧情提取钩子;建立“承诺—观众问题—风险—延期—兑现”契约。
  6. 把心理解释、作者评价和抽象氛围改成可见或可听证据。
  7. 先设计人物站位、路线、动作相位、道具和轴线,再设计摄影机。
  8. 标记必须看到、必须听到和可跨镜延续的事件。
  9. 建立 must-have Coverage 和临时 Shot Intent,再规划 Camera Setup。
  10. 冻结原子 Shot,计算动作、对白、反应、摄影机和声音的关键路径时长。
  11. 写入镜头间物理连续性状态,并执行叙事连贯性与镜头序列复核。
  12. 逐镜选择真人、Seedance 或混合制作,并设置备用方案和切换条件。
  13. 将每个 Seedance Shot 编译为 Generation Task;默认一镜一任务,多次 Attempt。
  14. 执行质量门禁;按责任阶段定向修复,最多三轮最小补丁。
  15. 输出分镜表、Seedance 任务表、差异报告、质量报告和风险/降级说明。

强制可拍规则

每个正式 Shot 默认必须同时满足:

  • 只有一个连续时空。
  • 只有一个主导景别;不得在同一镜号中同时写全景、近景和特写。
  • 只有一个主要视觉行动单元和一个主要叙事目的。
  • 只有一个主要摄影运动;复杂运动必须有完整路径和必要性。
  • 明确主体、开始状态、可见动作和结束状态。
  • 动作、对白、反应和运镜能在镜头时长内完成。
  • 能追溯到原文事实或明确授权。
  • 上一镜 end_state 能承接下一镜 start_state。

出现以下任一情况时拆镜:

  • 主导景别明显跨级。
  • 出现“切到、再切至、转到另一边”等隐含剪辑。
  • 机位、方向、注意力中心、时空或独立信息结果改变。
  • 反应或证据需要独立读取。
  • 动作链、主体数、口型、手部、物理交互或运镜负载过高。

只有用户明确需要连续长镜头且方案可执行时,才设置 long_take_exception: true,并写出起止构图、摄影机路径、人物路径、失败点和拆镜备用方案。

读取:

把空话变成画面

删除不能被摄像机或麦克风记录的解释,但保留其剧情信息。优先转译为:

  1. 有结果的身体动作。
  2. 人物与道具的状态变化。
  3. 人物距离、站位或空间关系变化。
  4. 明确、可重复表演的面部或手部变化。
  5. 环境的物理反应。
  6. 沉默、呼吸、脚步、物件声或必要对白。

不要写:

她终于不再隐忍,这个冷笑不是得意。

改写为:

她摘下工牌,压在辞职信上,把两样东西推到经理面前。

不得用“电影感、压迫感、宿命感、高级感”替代构图、光线、动作或声音。读取 abstract-to-visible.md。

保持人物和空间连续

建立并逐镜更新:

  • 人物左右位置、面朝方向、视线目标、姿势和动作相位。
  • 服装、头发、妆容、污渍、伤势和表演强度。
  • 道具位置、状态、持有手和移动路径。
  • 门窗、灯光、屏幕内容、天气、时间和轴线状态。
  • 对白、环境声、音乐和声画桥的跨镜状态。

先读取 script-analysis.md 与 story-bible.md,再读取 blocking-and-continuity.md。

继续开发并修复不连贯

用户要求继续开发后续场景、下一集或优化已有分镜,或者反馈“前后接不上、动作跳了、镜头有点散、节奏断了”时,读取 iterative-storyboard-refinement.md。

必须:

  1. 冻结最近一个已接受版本与用户原始反馈,不从空白状态重做。
  2. 按剧情信息、Scene/Beat、调度空间、Shot/剪辑、声音节奏、生成装配六层诊断最早根因。
  3. 对目标镜头同时检查前一镜和后一镜,比较 end_state、start_state、动作相位、画面方向、道具和声音。
  4. 从责任层实施最小补丁,并标记受影响的下游 Shot、Task、Asset 和 Clip。
  5. 后续 Scene 继承上一个相关 Scene 的出口快照、未兑现 Hook、人物知识状态和连续性锚点。
  6. 用同一证据窗口复核修复前后结果;默认最多三轮最小补丁。

不得把“不连贯”自动解释为需要新增人物动机、冲突、反转或剧情事实。需要改变核心因果时停在建议区并请求 story 授权。

设计短剧钩子、节奏和台词

  • 首屏尽快给出可见冲突、异常、证据、风险或未完成动作。
  • 钩子必须来自原剧情;需要新增剧情时升级到 story 并记录授权。
  • 每个钩子必须有兑现、合法延期或明确撤销,不得只堆悬念。
  • 爽点必须表现为权力、信息、目标或物理状态的可见变化。
  • 台词必须可说、可回应、可计时;不要让人物朗读作者解释。
  • 同期长对白、多人精确口型或复杂抢话优先真人拍摄或分离声音生产。

按题材和平台读取:

编译 Seedance 任务

不要把官方“支持某能力”写成稳定成功保证。先读取 model-registry.yaml,再根据实际入口、地区、账号和构建记录运行快照。

每个 Generation Task 至少包含:

  • 对应 Shot、时长、画幅和模型档案。
  • 角色、场景、服装、道具和参考素材 ID。
  • 首帧状态、单一主导景别、分段动作时间线和尾帧目标。
  • 摄影机状态或一个简单连续运动。
  • 对白、环境声和音效的时间点。
  • 连续性目标、负面约束、验收条件、风险分数和降级路径。

默认策略:

text
1 Atomic Shot → 1 Generation Task → 多个 Generation Attempt

多人长对白、复杂手部小道具、多阶段物理交互、复杂运镜和多镜头内部剪辑均提高风险。失败时依次减少主体、缩短动作、锁定机位、拆镜、分离声音、切换模型、改真人或混合制作。

读取:

权利、安全和来源

未知的素材权利、肖像、声音、未成年人同意、隐私、平台政策或地区限制不得被自动修复为“已通过”。把这些问题路由给人工审核。

读取 rights-safety-compliance.md 和 source-catalog.md。对快速变化的模型规格、价格和平台政策重新核验,不沿用过期结论。

使用模板和 Schema

优先复制而不是重建:

机器输出按 schema-index.md 与 references/schemas/ 中的 JSON Schema 组织。对象关系、字段名或枚举不确定时,读取对应 Schema,不自行创造近义字段。

执行审查梯队

按以下顺序审查;任一阻断项退回责任阶段:

  1. 结构审:对象、ID、Schema、引用和流程顺序。
  2. 人设与连续性审:身份、关系、外观、站位、轴线、道具和状态。
  3. 爽点与节奏审:冲突启动、信息差、钩子承诺、延期和兑现。
  4. 台词审:可说性、时长、回应关系、口型和声音生产。
  5. 格式审:字段、表格、模板、提示词和机器可读性。
  6. 一致性审:Source → Beat → Shot → Task → Attempt/Asset → Timeline。

运行:

powershell
node scripts/validate-all.mjs --input <project.json> --pretty --require-validators
node scripts/run-regression-tests.mjs
node scripts/build-repair-plan.mjs --report <quality-report.json> --project <project.json> --pretty
node scripts/estimate-cost.mjs --input <project.json> --pretty

不要自动应用修复计划。禁止自动修改锁定事实、授权、权利、隐私、同意和合规状态。

交付门禁

只在以下条件全部满足时声明“方案通过”:

  • 锁定事实保留率 100%,未授权新增事实为 0。
  • 所有抽象来源均已转为可见或可听证据,抽象语言审查无 Error/Blocker。
  • 正式 Shot 原子结构通过率 100%。
  • must-have Coverage、相邻镜头连续性、关键声音、台词测时、钩子与爽点因果全部闭环。
  • 每个 must-preserve Beat 均有镜头覆盖;不存在孤立 Shot、无触发结果、提前知情或未经授权调序;每对相邻镜头都有明确承接或合法转场。
  • 人物知识状态、当前目标、关系有效区间、说话方式和行为锚点无冲突。
  • 分镜、提示词、连续性、模型档案和制作方式互相一致。
  • 阻断级权利、安全、隐私和合规遗漏为 0。
  • 所有默认值、推断、调序、压缩和授权新增均出现在差异报告中。
  • 全部适用验证器的 Blocker 与 Error 均为 0。

缺少真实生成、真人走戏或剪辑装配证据时,只能声明“方案通过”,不得声明“成品发布通过”。

© zyz254009-crypto, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 94 other files (scripts, references, assets) in the repository root of zyz254009-crypto/script-to-shootable-storyboard.

  • SKILL.md
  • .gitignore
  • LICENSE
  • PROJECT_OVERVIEW.md
  • README.md
  • agents/openai.yaml
  • assets/existing-storyboard-repair-template.md
  • assets/generation-task-template.md
  • assets/project-input-template.yaml
  • assets/quality-report-template.md
  • assets/rights-manifest-template.yaml
  • assets/storyboard-template.csv
  • assets/storyboard-template.md
  • references/abstract-to-visible.md
  • references/atomic-shot-rules.md
  • references/blocking-and-continuity.md
  • references/cost-complexity.md
  • references/editing-and-coverage.md
  • … and 77 more

Open the folder on GitHubat commit e1ee698

Compare with similar skills

Script To Shootable Storyboard next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Script To Shootable Storyboard compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Script To Shootable Storyboard this skillzyz254009-crypto/script-to-shootable-storyboard117—~1.9kAutomated safety check: PassMIT
Seedance Storyboard Generatorliangdabiao/Seedance2-Storyboard-Generator2.6k—~2.2kAutomated safety check: PassNone
Cs Auto VidelChenShuo2004/cs-skills197—~2.1kAutomated safety check: PassMIT
Seedancefreestylefly/wesight946—~2.2kAutomated safety check: PassMIT
Higgsfield Motion DesignOSideMedia/higgsfield-ai-prompt-skill713—~2.3kAutomated safety check: PassMIT
Higgsfield CanvasOSideMedia/higgsfield-ai-prompt-skill713—~2.2kAutomated safety check: PassMIT

Similar skills

  • Seedance Storyboard Generator

    liangdabiao/Seedance2-Storyboard-Generator

    专业的Seedance 2.0平台AI视频脚本和分镜生成器。当用户要求:(1) 将文章/故事转换为视频脚本,(2) 生成Seedance 2.0分镜提示词,(3) 规划多集AI视频系列,(4) 为GPT-Image-2、Seedream、Nano Banana…

    2.6k GitHub stars~2.2k tokensUpdated 20 days ago
    Media & CreativeAuto-check passed
  • Cs Auto Videl

    ChenShuo2004/cs-skills

    A skill your agent uses when the user types $cs-auto-videl or asks for Douyin/TikTok ecommerce video workflows, including competitor replication, strong-Hook nine-grid promptimage.md to image2…

    197 GitHub stars~2.1k tokensUpdated 3 days ago
    Media & CreativeAuto-check passed
  • Seedance

    freestylefly/wesight

    Generate AI videos using Volcengine Seedance model. An agent skill from freestylefly/wesight.

    946 GitHub stars~2.2k tokensUpdated 10 days ago
    Media & CreativeAuto-check passed
  • Higgsfield Motion Design

    OSideMedia/higgsfield-ai-prompt-skill

    End-to-end motion-design / animated-ad creation flow on Higgsfield via the MCP connector.

    713 GitHub stars~2.3k tokensUpdated 14 days ago
    Media & CreativeAuto-check passed
  • Higgsfield Canvas

    OSideMedia/higgsfield-ai-prompt-skill

    A skill your agent uses when the user mentions Higgsfield Canvas, a node-based or node graph workspace, an infinite board/canvas, chaining generations into a pipeline, or wants to wire prompts →…

    713 GitHub stars~2.2k tokensUpdated 14 days ago
    Media & CreativeAuto-check passed
  • Seedance2 API

    aiskillstore/marketplace

    Out-of-the-box Seedance 2.0 API skill — just one API key to generate AI videos.

    433 GitHub stars~3.7k tokensUpdated yesterday
    Media & CreativeAuto-check passed

Works with

Questions about Script To Shootable Storyboard

What does Script To Shootable Storyboard do?

将剧本、故事梗概、对白稿、已有分镜或视频生成提示转换、检查并优化为可拍、可剪、适合 Seedance、真人或混合制作的原子分镜。用于剧本转分镜、分镜可拍化、后续场景开发、反馈驱动迭代、不连贯诊断与修复、心理描写视觉化、单镜主导景别拆分、短剧钩子与节奏、台词和声音时长、人物道具连续性、Seedance Generation Task、生成风险与失败降级;默认忠实转换,未获授权不得改写剧情。. Script To Shootable Storyboard is an agent skill from zyz254009-crypto/script-to-shootable-storyboard.

When should I use Script To Shootable Storyboard?

Script To Shootable Storyboard fits situations like: tasks that involve AI video generation; tasks that involve Comics and storyboards.

How do I install Script To Shootable Storyboard in Claude Code?

Run `npx skills add zyz254009-crypto/script-to-shootable-storyboard --skill script-to-shootable-storyboard -a claude-code`. Or copy the skill folder (the zyz254009-crypto/script-to-shootable-storyboard repository) into .claude/skills/script-to-shootable-storyboard in your project. Claude Code loads it when a task matches its description.

How do I install Script To Shootable Storyboard in Codex?

Run `npx skills add zyz254009-crypto/script-to-shootable-storyboard --skill script-to-shootable-storyboard -a codex`. Or copy the skill folder (the zyz254009-crypto/script-to-shootable-storyboard repository) into .agents/skills/script-to-shootable-storyboard in your project. Codex loads it when a task matches its description.

Can I use Script To Shootable Storyboard in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add zyz254009-crypto/script-to-shootable-storyboard --skill script-to-shootable-storyboard -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/script-to-shootable-storyboard, .gemini/skills/script-to-shootable-storyboard, .github/skills/script-to-shootable-storyboard and .opencode/skills/script-to-shootable-storyboard in your project.

What does Script To Shootable Storyboard need to run?

Going by SKILL.md and its folder, Script To Shootable Storyboard needs the command-line tools its instructions call (node).

Does Script To Shootable Storyboard access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Script To Shootable Storyboard safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Script To Shootable Storyboard use?

Script To Shootable Storyboard is published under the MIT licence (from the LICENSE file in the skill folder). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Script To Shootable Storyboard use?

About 1.9k tokens (SKILL.md is roughly 7.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 178k tokens, read only when the agent opens those files.

What are the alternatives to Script To Shootable Storyboard?

Skills that share tags, products or a category with Script To Shootable Storyboard: Seedance Storyboard Generator (liangdabiao/Seedance2-Storyboard-Generator, 2.6k stars), Cs Auto Videl (ChenShuo2004/cs-skills, 197 stars), Seedance (freestylefly/wesight, 946 stars) and Higgsfield Motion Design (OSideMedia/higgsfield-ai-prompt-skill, 713 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Script To Shootable Storyboard?

zyz254009-crypto (a GitHub user) maintains it in zyz254009-crypto/script-to-shootable-storyboard, which has 117 GitHub stars. The repository was last updated on July 17, 2026.

Source: zyz254009-crypto/script-to-shootable-storyboard on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.