spoken-script-polish
Skill29 starsChanged 7 months agoMITView on GitHub
Spoken-script polishing for Chinese videos, tutorials, and talking-head content. Use when Codex needs to turn a readable draft into something that can actually be spoken smoothly: remove written-language stiffness, shorten breath units, mark pauses or emphasis where useful, and preserve meaning while making the script sound natural aloud.
In this repository26 other files it gives its agents.
- ai-writing-diagnosisSkill
- benchmark-filterSkill
- book-material-minerSkill
- cc-designSkill
- celf-style-writerSkill
- content-strategy-diagnosisSkill
- deep-writerSkill
- hook-optimizerSkill
- horizontal-vertical-analysisSkill
- learning-builderSkill
- mrbeast-perspectiveSkill
- naval-perspectiveSkill
- opc-case-researchSkill
- paul-graham-perspectiveSkill
- perspective-distillerSkill
- short-video-script-builderSkill
- speech-structure-coachSkill
- steve-jobs-perspectiveSkill
- video-packaging-diagnosisSkill
- wechat-draft-publisherSkill
- x-article-publisherSkill
- x-content-mentorSkill
- xhs-image-directorSkill
- xhs-title-selectorSkill
- youmindSkill
- zhang-yiming-perspectiveSkill
The file
Files
Spoken Script Polish
Overview
这个 Skill 处理的不是“写什么”，而是“怎么说出来更顺”。
适合：
- 视频口播
- 教程解说
- Talking head
- 直播切片文稿
Quick Start
- 先判断脚本是信息不清，还是只是说起来不顺。
- 按三遍处理：内容、口语化、节奏标注。
- 一切改写都以大声读是否顺为标准。
Workflow
Step 1: Content Check
先确认：
- 事实是否正确
- 步骤是否连贯
- 时长是否合理
Step 2: Oral Rewrite
重点处理：
- 书面语
- 长句
- 过度完整的逻辑连接词
- 一看就像写出来而不是说出来的段落
Step 3: Delivery Pass
按需要标注：
- 停顿
- 重音
- 切屏
- 演示动作
Hard Rules
Do not:
- 只是机械加口头禅
- 为了自然牺牲信息准确度
- 把所有句子都改得碎碎的
Always:
- 以可说性为第一标准
- 缩短呼吸单位
- 保留原意和节奏
Resource Map
- references/oral-rhythm.md
- 口语化改写与节奏处理原则。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.