Individual skill
novel-script
给 AI 短剧写剧本:把 novel-outline 的分集梗概落成结构化的场次 + 节拍流(动作节拍与台词行交替), 台词逐句带说话人与语气,时长逐集按语速确定性折算。产出 script.json + Markdown + 单页评审报告 (时长仪表 / 分集剧本 / 场次总表 / 按角色聚合的台词本,含导出 JSON)。 剧本管戏,分镜管拍——本 skill 不分镜头、无镜号、不写生成提示词,那些是下一层分镜 skill 的活。 10 道质量门全部由脚本确定性检查(每集时长 ±15%、单句 ≤35 字、说话人合法、钩子悬念落纸、 钩子具象前 3 拍内兑现(hookBeat 认领冷开场)、每场至少一个动作节拍、爽点认领、 角色/场景/光照/道具对账上游……)。 有 outline.json 就用 seed 预填每集骨架;给 --art 连光照状态一起对账。 零依赖、零 API key,用当前会话额度。 Use when asked to 写剧本、出剧本、台词、场次、写戏、screenwriting for AI short drama。
skills/novel-script/SKILL.md
Source commit 5612d2a0bd5f19d978c3ee19d1e42840335e3b4b
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- Yes
- Repository scan complete
- No
- Skill scan complete
- No
- Observed behavior
- Network or broad shell
- Last checked
- Nov 9, 2026
These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.
Install this skill
Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.
An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.
Repository context
- Repository stars (not skill ratings)
- 6,327
- Repository last updated
- Oct 26, 2026
- License
- Apache-2.0
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- engineering Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
README.en.mddoc · 7,375 bytesREADME.mddoc · 7,605 bytesSKILL.mdcore · 9,204 bytesSKILL.mdcore · 9,204 bytesSKILL.mdcore · 9,204 bytesSKILL.mdcore · 9,204 bytesSKILL.mdcore · 9,204 bytesSKILL.mdcore · 9,204 bytesSKILL.mdcore · 9,204 bytesSKILL.mdcore · 9,204 bytesassets/report.webpasset · 106,104 bytesexamples/渡口-script.jsonasset · 38,592 bytesreferences/report-style.mddoc · 3,192 bytesreferences/schema.mddoc · 3,589 bytesreferences/script-pass.mddoc · 5,515 bytesreferences/test-fixtures/upstream/README.mddoc · 542 bytesreferences/test-fixtures/upstream/渡口-art.jsonasset · 16,839 bytesreferences/test-fixtures/upstream/渡口-cast.jsonasset · 42,544 bytesreferences/test-fixtures/upstream/渡口-outline.jsonasset · 9,659 bytesscripts/novel-script.mjsruns · 53,238 bytesscripts/selftest.mjsruns · 18,366 bytes
Read excerpt
## novel-script
给 AI 短剧写**剧本**。**前提刻在骨子里:剧本管戏,分镜管拍**——「爽不爽」和「怎么拍」是两种迭代节奏,台词要反复推翻重写,绑上镜头分解每改一句都得重排镜头。所以这层只有集、场次、节拍流,没有镜号;镜头、首帧提示词、生成批次都是下一层分镜 skill 的活。
但有一条底线:**台词必须是结构化数据,不能写成散文**。每句台词是独立条目(说话人 + 台词 + 语气),动作是独立节拍——这是全部确定性检查的地基:
| 交付 | 解决什么 |
| --- | --- |
| 逐集时长预算 | 一集三分钟就是三分钟:台词按语速折算、动作按节拍估时,写超写欠当场拦下,不流到生成环节才发现 |
| 节拍流(动作 ⇄ 台词) | 台词直接对接 TTS 逐句生成;动作节拍就是画面要发生的事。混成散文两头都喂不进管线 |
| 每场至少一个动作节拍 | 纯对白的场是广播剧——AI 生成时没有画面可写 |
| 开场钩子 + 结尾悬念 | 短剧的生死线,每集都要落在纸面;**钩子不是标签是第一拍**——`hookBeat` 认领具象位置,必须落在全集前 3 拍内(冷开场) |
| 爽点认领 | 大纲说这一集有大爆点,剧本必须有戏认领它——防止改着改着把爆点改丢 |
| 上游对账 | 场景/光照/道具对 art.json、角色对 outline.json,写了美术没登记的光照状态当场报 |