promptschat
Gives Claude Code access to the prompts.chat library, letting you search, retrieve and improve AI prompt templates, and find and install Agent Skills from its registry.
Turns a Bilibili or YouTube lecture or tutorial into a structured, figure-rich LaTeX course note and PDF, so you get study notes without rewatching the video. It also cleans up the video's subtitles and draws clear visuals for tensor and matrix formulas.
Claude Code
$ npx -y skills add wdkns/wdkns-skills --agent claude-code
Codex
$ npx skills add wdkns/wdkns-skills -a codex
Cursor
$ npx skills add wdkns/wdkns-skills -a cursor
Gemini CLI
$ npx skills add wdkns/wdkns-skills -a gemini-cli
GitHub Copilot
$ npx skills add wdkns/wdkns-skills -a github-copilot
OpenCode
$ npx skills add wdkns/wdkns-skills -a opencode
Windsurf
$ npx skills add wdkns/wdkns-skills -a windsurf
youtube-render-pdf:将 YouTube 讲座、教程或技术分享整理为结构化、配图丰富的中文 LaTeX 讲义,并编译交付最终 PDF。
bilibili-render-pdf:面向 Bilibili 视频生成中文 LaTeX 讲义和最终 PDF,并针对字幕缺失、登录高清与分 P 视频提供专门处理流程。subtitle-refine:在保持时间轴同步和原意不变的前提下精修 ASR 生成的中文 SRT 字幕,并通过校验脚本产出可发布的完整字幕。tensor-formula-viz:把张量、矩阵或向量公式及其代码路径转换为形状对齐、语义清晰且适合演示的计算示意图。这个仓库托管两个 Codex skill,用于将视频讲座转换为结构化的中文 LaTeX 讲义和最终 PDF。
| Skill | 平台 | 说明 |
|---|---|---|
youtube-render-pdf | YouTube | 原始版本,利用 YouTube CC 字幕和章节结构 |
bilibili-render-pdf | Bilibili (B站) | 适配 B 站的字幕缺失、登录高清、分P视频等特点 |
两个 skill 共享相同的写作规则、配图策略和 LaTeX 模板,但在素材获取阶段有平台特定的差异。
yt-dlp --cookies-from-browser chrome)whisper(openai-whisper)用于语音转写\section{} / \subsection{} 结构的完整 .tex如果你想在本地 Codex 环境中使用这些 skill,可以把对应目录放到你的技能目录中:
mkdir -p ~/.codex/skills
# YouTube 版
cp -R skills/youtube-render-pdf ~/.codex/skills/
# Bilibili 版
cp -R skills/bilibili-render-pdf ~/.codex/skills/
然后在 Codex 中使用对应 skill 处理视频链接,请求生成讲义 .tex 和最终 PDF。
| 工具 | 两个 skill 都需要 | 仅 Bilibili 版需要 |
|---|---|---|
yt-dlp | ✓ | |
ffmpeg | ✓ | |
xelatex (TeX Live + CTeX) | ✓ | |
magick (ImageMagick) | ✓ | |
whisper (openai-whisper) | ✓ |
此外,运行 skill 的 coding agent 必须具备一定的读图能力,否则很难选择关键帧,很难做到图文align(即至少是一个还不错的 vlm model,ps. MiniMax 2.7 只是一个纯文本模型)。
spwan_agent 的触发,规定的比较死,"Only use spawn_agent if and only if the user explicitly asks for sub-agents, delegation, or parallel agent work.",即需要我们在 query 中显式地要求,才可以触发 subagents$youtube-render-pdf https://www.youtube.com/watch?v=vXb2QYOUzl4 请 spwan 多 sub agents 执行,隔离上下文,避免 master agent 的“上下文焦虑”, 形成一个完整全面的 pdf:
- 1 个 outline agent:先定全局目录、术语、符号表、章节边界等
- 5 个 writer agents:各自直接写成完整章节草稿,落盘成 section_*.tex
- 1 个 figure agent:单独负责抽帧、筛图、crop、脚本生成新的示意图、图注和时间脚注等;
- 1 个 consistency agent:检查重复定义、前后术语不一致、章节衔接断裂
spwan 一个独立的reviewer agent,基于原始字幕文件,check 是否有重要、细节及有趣等一切有意的信息的漏召回,仅反馈,不修改,不断交互,直至reviewer agent 觉得 tex 信息已完备。
仓库保留了根目录下原有的 LICENSE 文件。使用、分发或二次修改时,请以该许可证为准。
Gives Claude Code access to the prompts.chat library, letting you search, retrieve and improve AI prompt templates, and find and install Agent Skills from its registry.
A Claude Code skill for generating editable PPTX presentations and slides from a workflow, with reusable brand, style, layout, and deck workspaces, native charts, speaker notes, and template filling.
Gives AI agents control over Word, Excel and PowerPoint files, with skills for building .docx reports, academic papers, Excel dashboards and Morph-style animated slides.
A skill that builds animation-rich HTML presentations from scratch or converts PowerPoint files to web slides. Useful for anyone turning a deck into a web presentation.
Install it with the command on its page. It includes bilibili-render-pdf, subtitle-refine, tensor-formula-viz. Its skills do not fire on their own yet. Free and open source.
Is this plugin yours?
Claim it with GitHubSubmit a pluginPromote it