Sketch Narrator

从中文主题、文案、视频链接、本地音视频或参考角色制作带自然配音、真实音频时间、分区笔迹和内部成片 QA 的完整手绘讲解视频。适用于素材转口播、黑板讲解、白板科普、故事口播和系列 IP 手绘视频;不用于纯 AI 视频模型生成或复杂 Web 工作台建设。

22 stars 5 forksTesting & QA

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add qingyabailu-git/SketchNarrator --skill sketch-narrator
Manual install
git clone --depth 1 https://github.com/qingyabailu-git/SketchNarrator.git /tmp/SketchNarrator
mkdir -p ~/.claude/skills
cp -r /tmp/SketchNarrator/. ~/.claude/skills/sketch-narrator

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# SketchNarrator 手绘讲解视频

用户提供主题、时长、画幅、可选风格和参考图。Codex 负责风格推荐、脚本、默认 Edge 或用户明确选择的可选配音、分镜、视觉编排、整板图、语义标注、渲染和验收。用户进行三次轻确认,每次只确认即将成为下游输入的内容:

1. 最终口播稿与项目风格。确认前禁止生成配音。
2. Edge 配音、字幕、分镜摘要、每幕文字卡片、中文视觉编排表、节奏与素材摘要。确认前禁止生成整板图。
3. 整板画面与绘制顺序。确认后才允许正式渲染。

不要要求用户复制提示词、运行命令、安装依赖或管理文件。内部 QA 不增加用户确认关卡。

## 宿主与安装准备

支持 Codex、Antigravity、WorkBuddy、豆包中具备本地代理能力的使用方式,共用本 Skill 的文件和确认流程。下文的 Codex 指执行任务的宿主代理,不限定产品名称。宿主必须能读写项目、执行本地命令、生成并查看图片、访问本机工作台;仅聊天或仅上传文档的界面不能独立完成本地生产。不得因宿主名称自动认定这些能力可用。

安装时先运行 `scripts/run.cmd doctor --host <codex|antigravity|workbuddy|doubao> --project <项目目录>`(macOS/Linux 使用 `sh scripts/run.sh`)。该入口不安装依赖、不请求在线服务;输出 Python、依赖、字体、FFmpeg 和目录权限预检。宿主能力由代理根据实际工具确认,可用 `--capability` 声明,但报告中的声明不是运行验证。未探测项保持 unknown,不能以命令退出成功宣称安装完成。

先准备缺少的核心依赖,再复查;FFmpeg 缺失时,在已获安装授权的范围内执行 `setup --install-ffmpeg`,没有授权才一次性询问。该步骤检查实际编码能力,不修改系统 PATH,不新增创作确认关卡。安装准备属于使用前的独立阶段,不授权生产中修改源码或擅自升级环境。没有兼容 Python 时停止并说明环境前提,不自动安装系统 Python。

## 公共运行入口

Python 与依赖采用最低版本要求,不预设未来版本上限。新环境由 pip 按当前 Python、平台和依赖元数据解析可用稳定版本;旧 Python 可以选择仍兼容的依赖版本,不强迫所有平台使用同一最新版。现有可用环境默认复用,只有明确调用 `setup --upgrade` 才请求升级核心及所选可选依赖。安装后检查导入与依赖一致性并记录实际版本;这些检查不代替视频验收。尚未运行的新版组合标为未验证,不能承诺未来版本永远兼容。只有可复现的不兼容问题才添加针对性版本排除,并记录原因与解除条件。

不得假设 `python`、`uv` 或任何用户目录中的解释器位于 PATH,也不得把开发者机器的绝对路径写入项目。Windows 调用 `scripts\run.cmd <子命令>`,macOS/Linux 调用 `sh scripts/run.sh <子命令>`;两者会优先复用仓库内隔离环境并在首次使用时通过系统 Python 3.10+ 引导依赖,低于最低 Python 要求或损坏的旧环境不得继续复用。下文的 `python scripts/workflow.py` 表示该跨平台入口对应的工作流子命令,不是要求用户手动执行。
Read the full skill on GitHub

Details

Skill name
sketch-narrator
License
MIT
Repo updated
30 September 2026

FAQ

What does the Sketch Narrator skill do?

从中文主题、文案、视频链接、本地音视频或参考角色制作带自然配音、真实音频时间、分区笔迹和内部成片 QA 的完整手绘讲解视频。适用于素材转口播、黑板讲解、白板科普、故事口播和系列 IP 手绘视频;不用于纯 AI 视频模型生成或复杂 Web 工作台建设。

How do I install the Sketch Narrator skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add qingyabailu-git/SketchNarrator --skill sketch-narrator. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Sketch Narrator popular?

It lives in qingyabailu-git/SketchNarrator, which has 22 GitHub stars and 5 forks.

Related skills

Zod

24

anivar/zod-skill

Zod v4 best practices, patterns, and API guidance for schema validation, parsing, error handling, and type inference in TypeScript applications. Covers safeParse, object composition, refinements, transforms, codecs, branded types, v3→v4 migration, and testing schemas with Jest or Vitest. Baseline: zod ^4.6.0. Triggers on: zod imports, z.object, z.string, z.infer, safeParse, mentions of "zod", "schema validation", "zod v4", "z.enum", or testing a schema — "schema test", "zod mock", zod-schema-faker.

Testing & QA

Molt Dex

23

ReSo7200/molt-dex

Reverse-engineer and assess Android apps, from unobfuscated to heavily obfuscated, using a staged, evidence-first pipeline built on apktool, adb and scrcpy. Use when the user wants to understand how an APK is built, what it does, where its secrets and misconfigurations are, or wants to drive an app on a device to observe its behaviour. Dynamic testing (anything that touches a running system or backend) is authorization-gated and must be in scope.

Testing & QA

Concept Reel

25

Jogzh/concept-reel

使用 Canvas、Playwright 和 FFmpeg 制作或修改无旁白的横屏字幕知识短片,支持 15 种动态设计风格、内容驱动分镜、确定性动画、预览与音效合成。适用于把概念或机制做成可理解的动画;有人声口播、实拍和生成模型视频不属于这套引擎的工作流。

Testing & QA

Premortem

30

b1rdmania/claude-premortem-skill

Run a premortem on any plan, launch, product, hire, strategy, or decision. Imagines it failed 6 months from now, works backward to find every reason why, then produces a revised plan. Triggers include "premortem this", "premortem my", "what could kill this", "stress test this plan", "find the blind spots", "poke holes in this", "where will this break", "am I missing anything", "what could go wrong", "future-proof this", "devil's advocate this".

Testing & QA

State of Paywalls 2026 By Adapty

26

vkharlan-adapt/state-of-paywalls-2026-skill

Answer questions from Adapty's State of Paywalls 2026 report and audit paywall designs against it: benchmarks by region and category, pricing, placement, A/B test ideas, and paywall screenshot reviews.

Testing & QA

Vibe Testing

24

knot0-com/vibe-testing

This skill should be used when the user asks to "test my specs", "validate my design docs", "find gaps in my architecture", "stress-test the spec", "vibe test", "pressure test the docs", or mentions spec validation before implementation begins.

Testing & QAIn 1 list

Browse the most popular Claude skills, all testing & qa skills, or write your own with the skills generator.