短剧剧本skill

SKILL内容创作
内容创作 未标注权限 许可证不明确
安装这个技能

支持热播短剧研究、原创选题评分、钩子与反转设计、真人/动漫、横竖屏、15 秒时间线、Word 生产稿及逐集导入包

一句话选型结论

按资料字段整理

短剧剧本skill是内容创作类的 Agent Skill,安装方式未标注,权限级别未标注,许可证不明确,有可直接复制的安装命令。技能本质是给模型读的说明书,效果取决于模型与客户端;装完先用一个小任务验证,再放进正式流程。

分类内容创作 安装方式未标注 权限级别未标注权限 许可证许可证不明确
累计安装 未标注 本周 未标注
GitHub 星标 未标注 仓库 未标注
下载量 未标注 技能市场下载数
权限级别 未标注权限 安装前请确认风险范围
许可证 许可证不明确 未说明

安装

未标注
在支持 Skills 的客户端里执行,装完问一句「你有哪些技能」验证
skillhub install user_5d448139/short-drama-ai-reels-skill
  1. 1
    确认客户端支持

    确认你的客户端支持 Skills(Claude、Codex、Cursor 等),并升级到较新的版本。

  2. 2
    拿到技能文件

    执行上面的命令安装;也可以把仓库里的 SKILL.md 手动放进客户端的 skills 目录。

  3. 3
    刷新技能列表

    重新打开客户端,或刷新技能列表,让新放进来的技能被加载。

  4. 4
    验证是否生效

    问一句「你有哪些技能」,确认列表里出现它,再拿一个小任务试跑。

技能档案

内容创作
字段来自技能仓库与公开索引,中文说明由本站标注
作者 / 维护方user_5d448139仓库 未标注 · 版本 未标注
宿主客户端未标注技能可被这些客户端加载
安装方式未标注安装包 community
权限级别未标注权限安装前请确认风险范围
许可证未说明许可证不明确,商用前请自行确认条款
仓库可用性available技能仓库当前是否可访问
最近核验未标注上游最后核验时间 · 数据更新 1791728196
正文体量6.3 KBSKILL.md 原文大小,越大说明说明书写得越细

典型使用场景

3 类
按分类整理的常见用法,具体以 SKILL.md 为准
写文章与脚本

按主题产出初稿与分镜脚本。

改稿与润色

把草稿改顺、统一语气。

多平台文案改写

同一内容按平台改成长短不同的版本。

选择建议

装之前先看这几条
按条目字段给出的判断,不替你决定用不用

值得选的地方

  • 有可直接复制的安装命令,不用自己翻文档找入口。

先想清楚的地方

  • 没有标注权限级别,实际能做什么要装好后确认。
  • 许可证不明确,商用或二次分发前先到仓库确认条款。
  • 没有做站内核验,内容以作者仓库为准。
  • 技能只是说明书,实际效果取决于模型与客户端,装完先用小任务试一遍。

更适合

  • 需要在内容创作场景里给模型补一套做法的人。
  • 愿意先读一遍 SKILL.md、确认它会让模型做什么的人。

不太适合

  • 对许可证要求很严、必须明确可商用的人。
  • 只想装完就自动变强、不打算看说明书的人。

同类对比

6 个同类
同分类里挑最接近的几条,字段以各自详情页为准
技能权限级别安装方式累计安装宿主客户端安装命令
短剧剧本skill当前 未标注权限 未标注 未标注 未标注 有
AI 视频生成 调用外部服务 GitHub 仓库 431.3 万 Codex、Claude、Cursor 有
视频帧 联网访问 ClawHub 市场 262.1 万 OpenClaw 有
Markdown 转换器 写入文件 ClawHub 市场 135.3 万 OpenClaw 有
哔哩哔哩全合一 接触敏感数据 ClawHub 市场 64.1 万 OpenClaw 有
Remotion video toolkit 命令行 浏览器控制 ClawHub 市场 63.9 万 OpenClaw 有

SKILL.md 全文

6.3 KB
模型实际读到的说明书原文,来自技能仓库

Short Drama and AI Reels Production

Create an original, high-retention short drama and move it from research to production without losing narrative or visual continuity. Support any positive episode count requested by the user; never impose a fixed total or an artificial maximum.

Start Here

Identify the user's current job and enter at that stage. Do not make them study the whole workflow first.

User request Start with Read only when needed
"查热播、分析爆款" Research and synthesis references/hit-analysis.md
"帮我选题、判断能不能火" Candidate scorecard and topic decision Hit analysis, then references/script-design.md
"写剧本、续写、重写" Series bible, arc, and episode drafting Script design
"改成真人/动漫、横屏/竖屏" Production adaptation references/ai-reels-production.md
"检查剧本、拆分、打包" Run the audit and packaging scripts references/quickstart-examples.md
"直接操作 AI Reels" Inspect the current UI and resume from its actual state AI Reels production and references/configuration-troubleshooting.md

For copy-ready prompts, a worked example, and common questions, read references/quickstart-examples.md. For parameter choices or failures, read references/configuration-troubleshooting.md.

Preserve User Intent

Honor the requested genre, exclusions, live-action versus animation, orientation, language, duration, model, platform, and episode count. If the user gives no count, choose a story-appropriate count or present a concise range based on the episode engine; do not stretch an idea merely to fill a preset number.

Do not force reincarnation, domineering-CEO romance, identity concealment, horror, or revenge formulas merely because they are common. Novelty should come from a simple explainable premise, a fresh social or emotional contradiction, and a repeatable episode engine.

End-to-End Workflow

  1. Define the research scope and metric meanings. Do not claim to have found "all" high-play dramas when platforms lack a unified public database.
  2. Extract transferable techniques instead of copying plots, dialogue, signature scenes, names, or relationship structures.
  3. Score several distinct premises for instant comprehension, emotional stakes, episode engine, reversal capacity, differentiation, and production feasibility.
  4. Select one premise and lock the series bible: logline, theme, protagonist, antagonistic force, relationships, reveal ladder, fixed assets, and ending.
  5. Scale the season architecture proportionally to the requested total. Every episode must change knowledge, power, relationship, possession, safety, reputation, or deadline.
  6. Draft production-ready episodes with stable character, scene, wardrobe, and prop identifiers. Each episode needs an opening disruption, objective, escalation, meaningful reversal/payoff, emotional beat, and final hook.
  7. Run scripts/audit_production.py. Fix errors; review warnings rather than blindly suppressing them.
  8. Create the Word production稿 and run scripts/package_episodes.py for per-episode text files and ZIP packaging.
  9. When asked to operate a website, verify the current model, style, orientation, episode count, prompt mode, duration, attached script, and credit implications before submission.

Quality Bar

  • The premise is understandable in one sentence and demonstrated visually near the beginning.
  • The protagonist makes consequential choices; coincidence may trigger the story but must not solve it.
  • Reversals change meaning, allegiance, cost, or power. Withholding an obvious fact is not a reversal.
  • Emotional depth comes from incompatible values and accumulated consequences, not long speeches.
  • Fast pacing means high information density and irreversible decisions, not constant shouting.
  • Reusable locations and fixed wardrobe improve generated-video consistency without making episodes visually static.
  • The prior episode's final hook is acknowledged promptly in the next episode.
  • Never promise virality. Give an evidence-based potential assessment and identify audience and execution risks.

Deliverables

For a complete project, normally produce:

  1. Research and topic-selection report with dated sources and metric caveats.
  2. Full production script in .docx with navigation-friendly headings.
  3. Import ZIP containing global rules, asset libraries, prompt templates, a manifest, and one UTF-8 text file per episode.
  4. Audit report showing episode continuity, scene structure, title uniqueness, hook markers, and packaging integrity.

Validate numbering, title order, scene presence, ZIP CRC, and Word heading structure. Render Word pages when the document toolchain is available; disclose clearly when visual rendering cannot be performed.

Failure Handling

Return actionable recovery steps, not only an error message. For every blocker, state:

  • What succeeded
  • The exact failing stage
  • Likely cause supported by evidence
  • What the user can change
  • What the skill will retry
  • The stopping condition that prevents duplicate projects or credit consumption

Use the detailed recovery table in references/configuration-troubleshooting.md. Do not repeat the same submission after the same server response.

External Actions

Uploading files, creating projects, or consuming generation credits are separate external actions. Preserve narrow authorization already given, but confirm immediately before a new credit-consuming generation when authorization is absent or ambiguous. A request to write a script does not authorize uploading or spending credits.

常见问题

5 问
答案由本站依据条目字段整理
短剧剧本skill怎么安装?

在支持 Skills 的客户端里执行:skillhub install user_5d448139/short-drama-ai-reels-skill。安装方式为未标注。装完刷新技能列表或重开客户端。

短剧剧本skill支持哪些客户端?

资料没有标注宿主客户端,但只要客户端支持 Skills(Claude、Codex、Cursor 等),一般都能加载。

短剧剧本skill权限高吗,安全吗?

权限级别为未标注权限。安装前先看它会读取或写入哪些内容。

短剧剧本skill收费吗?

本站不代收费用,也没有标注具体价格。是否收费取决于技能本身和所用的模型或客户端,正式使用前按作者说明确认。

短剧剧本skill用什么许可证?

许可证标注为 未说明(许可证不明确)。商用或二次分发前请到仓库确认条款原文。

热门专题

按场景挑技能
全部专题 →

技能内容以作者仓库为准;分类、权限级别与许可证状态的中文说明为中文整理,最近核验 未标注。