sugarforever/boring-video-studio

sugarforever 资料

最近记录的来源活动: 2026年10月11日 20:31

SkillsMP 收录数据更新: 2026年10月17日 00:09

已收集 skills: 12

GitHub 星标: 174
GitHub Forks: 26

仓库 README

仓库 README 来源：README.md

查看 README 源文件

展开完整 README

# Boring Video Studio Boring Video Studio is a collection of [Agent Skills](https://agentskills.io) for planning high-quality videos before production. For content-driven work, the main `boring-video` skill establishes and approves the content before it plans the video: ```text content/TOPIC.md → content/ARTICLE.md → BEATS.md → NARRATION.md → SCENES.md → STORYBOARD.md ``` Research notes and experiment reports live beside the content contracts under `content/`. Product promos, UI demonstrations, music-led pieces, locked scripts, and other direct-production inputs can bypass the article chain and enter HyperFrames through its own intake. It then hands the approved plan to [HyperFrames](https://www.hyperframes.dev), which owns video production, review, timing, animation, verification, and rendering. Optional [ListenHub](https://listenhub.ai) TTS support can provide narration audio and subtitles through the `listenhub-tts` skill. The existing `verysmallwoods-video` skill remains available as the complete VerySmallWoods publishing workflow for videos, covers, platform copy, blog posts, and social posts. ## Installation Install `boring-video`, its planning stages, and the setup verifier: ```bash npx skills add sugarforever/boring-video-studio \ --skill setup-boring-video-skills boring-video develop-topic to-article to-spec-beats to-narration to-scenes to-storyboard to-video record-terminal ``` Add `--global` to make the skills available across projects: ```bash npx skills add sugarforever/boring-video-studio --global \ --skill setup-boring-video-skills boring-video develop-topic to-article to-spec-beats to-narration to-scenes to-storyboard to-video record-terminal ``` The repository groups these coordinated skills under `skills/boring-video/`, but installation exposes their existing names as a flat skill set. `verysmallwoods-video` stays at `skills/verysmallwoods-video/` and can be installed independently. After installation, invoke `$setup-boring-video-skills` once to verify the complete set and run the HyperFrames environment check. Video production requires HyperFrames and its local dependencies. Check the environment with: ```bash npx hyperframes doctor ``` ## Usage Invoke `$boring-video` with a topic or source document: ```text $boring-video Turn this article into a planned eight-minute faceless explainer. ``` For a content-driven video, the skill asks for any missing production choices and runs the planning chain: 1. `develop-topic` researches the topic, records experiments, and obtains approval for the thesis, argument, evidence, and boundaries. 2. `to-article` writes the platform-independent content master and obtains article approval. 3. `to-spec-beats` reorganizes the approved article into the video's learning progression. 4. `to-narration` writes the spoken draft without adding new content. 5. `to-scenes` designs visual events, worlds, action, and continuity. 6. `to-storyboard` translates the scenes into buildable shots. 7. `to-video` hands the complete plan and optional ListenHub narration to the appropriate HyperFrames workflow. Every planning stage maintains one authoritative Markdown artifact at a stable path. Codex opens that file directly for reading and uses the source or diff review surface for line comments; Git holds its history. Rendered images and media remain evidence referenced from Markdown, while HyperFrames Studio owns playable timeline and motion review. An existing complete article can start at `to-spec-beats` after the user explicitly confirms it as the content master. Later artifacts are resumable only when their lineage traces to that approved article. You can also invoke an individual stage when you already have its input: Install an individual stage together with the Skills it invokes. For example, add `record-terminal` when the storyboard contains terminal, CLI, database, or TUI shots: ```bash npx skills add sugarforever/boring-video-studio --skill record-terminal to-storyboard ``` ```text $to-storyboard Turn SCENES.md into a HyperFrames storyboard. ```

正在加载仓库 README…

Boring Video Studio：先规划内容，再进入视频制作

把主题或已有文章逐层整理成确认过的内容、口播、视觉场景和 storyboard，再交给 HyperFrames 制作视频。

用途

内容型视频按选题与证据、内容母稿、学习节奏、口播、场景和镜头划分责任。产品宣传、UI 演示、音乐主导的视频和已锁定脚本可直接进入 HyperFrames，不必先写文章。

必要前提

使用本次发布介绍的重组套件时，安装后先运行 $setup-boring-video-skills，检查协同 Skills 与 HyperFrames 环境。实际视频制作需要 HyperFrames 及其本地依赖；README 给出的环境检查命令是 npx hyperframes doctor。ListenHub 配音属于可选分支，使用独立的 listenhub-tts Skill。

操作说明

原帖给出的起步请求是 $boring-video 把这个主题制作成一条八分钟的 faceless explainer。提供主题或源文档；已有文章需先确认是内容母稿。主流程依次生成 content/TOPIC.md → content/ARTICLE.md → BEATS.md → NARRATION.md → SCENES.md → STORYBOARD.md，并在权威 Markdown 文件旁生成 HTML 审核预览。已有确认文章可以从 to-spec-beats 开始；已有场景时可用原帖的 $to-storyboard 把 SCENES.md 转换成可制作的 HyperFrames storyboard。

限制

这里描述仓库本次重组后的版本。规划层确定内容与视觉意图；HyperFrames 负责实现、时间线、动画、审核、验证和渲染。恢复后续产物时，必须能追溯到确认过的文章。已锁定脚本或直接制作类输入不需要先虚构一篇文章。

引用来源

这个仓库中的 skills

职业分类待补全

未分类

小木头的个人视频流水线。基于一个主题，一篇文章，或一份口播稿，生成一整套可发布物料：视频成片，封面图，YouTube/Bilibili 文案，博客，以及社交媒体推广文案。视频制作采用 HyperFrames。

2026-10-09

已展示 12 / 12 个已收集 Skill。

用 AI 助手安装

复制这段 Prompt，发给你正在使用的 AI 助手。

Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/sugarforever/boring-video-studio.

sugarforever/boring-video-studio

仓库 README

Boring Video Studio：先规划内容，再进入视频制作

用途

必要前提

操作说明

限制

相关分享

这个仓库中的 skills

用 AI 助手安装

仓库 README