creating-explainer-videos-skill 2.0.0__tar.gz

creating_explainer_videos_skill-2.0.0/.gitignore

@@ -0,0 +1,8 @@
+node_modules/
+dist/
+.worktrees/
+*.tgz
+__pycache__/
+*.pyc
+.DS_Store
+Thumbs.db

creating_explainer_videos_skill-2.0.0/CHANGELOG.md

@@ -0,0 +1,34 @@
+# Changelog
+
+All notable changes are documented here.
+
+## [2.0.0] - 2026-08-31
+
+### Changed
+
+- Renamed the product from an AI-series-specific Skill to the domain-neutral `creating-explainer-videos-skill` package and `creating-explainer-videos` Skill.
+- Promoted `explainer-video-skill` to the primary CLI; kept `ai-principle-video-skill` as a deprecated compatibility alias.
+- Moved series-specific defaults into the isolated `ai-principle-series` example preset.
+
+### Added
+
+- Executable production state machine and JSON-first project lifecycle commands.
+- Runnable renderer and dedicated cover generated for every new project.
+- Canonical narration normalization, escaped/snake-case underscore guard, caption/TTS identity, and measured timing import.
+- Deterministic render, mux, ffprobe/FFmpeg audit, artifact hashing, packaging, and human-release decision boundary.
+- Three-template visual collection: Paper Theatre, Spatial Chamber, and Ink Explainer.
+- Spatial Chamber depth/path grammar, licensed-plugin fallbacks, and QC limits derived from the former B visual direction.
+- Declarative presets, permission-listed extensions, immutable hash metadata, and legacy research ID alias.
+- Credit-card clearing and quantum-tunneling non-AI fixtures.
+- Windows/Linux and Node.js 18/22 CI plus packed npx/global install smoke tests.
+- PyPI wheel/source distribution with a dependency-free Python lifecycle CLI, bundled canonical Skill assets, Node.js runtime delegation, archive security audit, and clean-environment smoke test.
+
+### Security and licensing
+
+- No font, premium GSAP, music, audio, image, or video binaries are redistributed.
+- Automated QC cannot set `passed`; a named human must completely review the exact artifact hash.
+- Package commands use argument arrays and keep provider credentials outside projects and logs.
+
+## [1.1.0] - 2026-08-30
+
+Legacy AI-series-specific first public release: portable installer, declarative extension API, integrity manifest, Ink/Paper profiles, episode scaffold, and release gates.

creating_explainer_videos_skill-2.0.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 wzwailr
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

creating_explainer_videos_skill-2.0.0/PKG-INFO

@@ -0,0 +1,180 @@
+Metadata-Version: 2.5
+Name: creating-explainer-videos-skill
+Version: 2.0.0
+Summary: Executable Agent Skill and scaffold for producing rigorous mechanism and process explainer videos.
+Project-URL: Homepage, https://github.com/wzwailr/creating-explainer-videos-skill
+Project-URL: Repository, https://github.com/wzwailr/creating-explainer-videos-skill
+Project-URL: Issues, https://github.com/wzwailr/creating-explainer-videos-skill/issues
+Project-URL: Changelog, https://github.com/wzwailr/creating-explainer-videos-skill/blob/main/CHANGELOG.md
+Author: wzwailr
+License-Expression: MIT
+License-File: LICENSE
+Keywords: agent-skill,educational-video,explainer-video,gsap,motion-design
+Classifier: Development Status :: 4 - Beta
+Classifier: Environment :: Console
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Topic :: Multimedia :: Video
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+
+# Explainer Video Skill
+
+[![CI](https://github.com/wzwailr/creating-explainer-videos-skill/actions/workflows/ci.yml/badge.svg)](https://github.com/wzwailr/creating-explainer-videos-skill/actions/workflows/ci.yml)
+[![PyPI](https://img.shields.io/pypi/v/creating-explainer-videos-skill)](https://pypi.org/project/creating-explainer-videos-skill/)
+[![Release](https://img.shields.io/github/v/release/wzwailr/creating-explainer-videos-skill)](https://github.com/wzwailr/creating-explainer-videos-skill/releases)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
+
+一套通用、可安装、可扩展、真正可执行的解释型视频 Agent Skill 与工程脚手架。它适用于机制、流程、系统、科学和技术科普，不绑定某个系列或某类题材。
+
+它不只是几份 Markdown：包内包含生产状态机、项目生成器、旁白规范化、真实音频时轴导入、三套可编程视觉模板、确定性 HTML/SVG/GSAP 渲染器、HyperFrames/FFmpeg 适配、媒体质检、扩展 API、打包工具和非 AI 端到端示例。
+
+[English](README.en.md) · [v2 迁移指南](docs/MIGRATION_V2.md) · [视觉模板集](skill/creating-explainer-videos/references/visual-template-collection.md) · [扩展 API](skill/creating-explainer-videos/references/extension-api.md)
+
+## 一键使用
+
+推荐使用 `pipx` 安装独立 CLI：
+
+```powershell
+pipx install creating-explainer-videos-skill==2.0.0
+explainer-video-skill --version
+```
+
+也可以安装到当前 Python 环境：
+
+```powershell
+python -m pip install creating-explainer-videos-skill==2.0.0
+explainer-video-skill --version
+```
+
+Python 入口可在没有 Node.js 时完成 Skill 的安装、更新、校验、回滚、卸载和扩展列表。创建视频工程、模板预览、渲染和媒体命令使用包内同一套 JavaScript runtime，需要 Node.js 18+。
+
+需要 npm 形式时，可直接安装 GitHub Release 中经过验证的 tarball：
+
+```powershell
+npm install --global https://github.com/wzwailr/creating-explainer-videos-skill/releases/download/v2.0.0/creating-explainer-videos-skill-2.0.0.tgz
+```
+
+安装到 Codex：
+
+```powershell
+explainer-video-skill install --target codex --json
+explainer-video-skill verify --target codex --json
+```
+
+安装到任何支持目录型 `SKILL.md` 的 Agent、仓库或团队目录：
+
+```powershell
+explainer-video-skill install --destination "C:\path\to\skills" --json
+explainer-video-skill verify --destination "C:\path\to\skills" --json
+```
+
+`--destination` 指向 skills 父目录，实际安装目录是 `creating-explainer-videos`。升级、回滚和卸载均保留可恢复备份。
+
+## 直接创建视频工程
+
+```powershell
+explainer-video-skill doctor --json
+explainer-video-skill templates list --json
+explainer-video-skill new ".\my-video" `
+  --title "信用卡清算" `
+  --topic "为什么授权成功后钱还没有到账" `
+  --template spatial-chamber `
+  --preset general-mechanism `
+  --json
+explainer-video-skill status --json ".\my-video"
+explainer-video-skill next --json ".\my-video"
+```
+
+工程会生成 JSON-first 生产合同、状态机、可运行 renderer、独立 cover、模板资产、旁白/cue 文件、媒体/QC/发布目录。Agent 按以下闭环推进：
+
+```text
+status -> next -> 执行真实工作 -> 写入证据 -> validate -> 下一阶段
+```
+
+核心阶段：
+
+```text
+brief -> evidence -> mechanism_map -> narration_and_cues
+-> real_audio_timing -> scene_spec -> runnable_renderer
+-> render -> automated_qc -> human_listen
+-> publishing_package -> human_release_decision
+```
+
+自动化只能生成 `release_candidate_pending_human_listen`。只有人工完整听看过同一个 SHA-256 成片并明确记录决定，才能成为 `passed`。
+
+## 视觉模板集
+
+| 模板 | 最适合的知识结构 | 核心动作 |
+| --- | --- | --- |
+| `paper-theatre` | 对象身份、证据、分拣、比较、规则、物理处理 | 剪、折、盖章、堆叠、交接 |
+| `spatial-chamber` | 路由、层级、管线、队列、网络、状态迁移 | 镜头纵深、路径飞行、分层变换 |
+| `ink-explainer` | 推导、公式、因果关系、比较、纠错、密集讲解 | 手绘、连接、圈注、推导、修正 |
+
+三者是结构和运动语法不同的模板，不是 A/B/C 换色版。每套都包含独立 DOM 指纹、场景 CSS、封面 CSS、时间轴控制器、插件降级和可量化 QC 规则。Spatial Chamber 正式收录了原 B 方案的纵深、路径和空间叙事能力。
+
+钉钉进步体/DingTalk Sans 可作为本机显示字体，Noto Sans SC 可作为字幕字体；仓库不分发字体文件。GSAP 商业插件同样不随包分发，每个模板都声明了开源/原生降级路径。
+
+## 旁白、动画与渲染
+
+字幕与 TTS 来自同一个 canonical 字符串。规范化会移除 `\_`、snake_case 下划线和可能被读成符号名的 Markdown 标记。真实旁白生成后导入测量时轴：
+
+```powershell
+explainer-video-skill narration prepare ".\my-video" --json
+explainer-video-skill narration import-timing ".\my-video" --timing ".\timing.json" --json
+explainer-video-skill build ".\my-video" --json
+explainer-video-skill render ".\my-video" --json
+explainer-video-skill cover ".\my-video" --json
+explainer-video-skill mux ".\my-video" --audio ".\narration.wav" --json
+explainer-video-skill audit ".\my-video" --json
+explainer-video-skill package ".\my-video" --json
+```
+
+最终动画只使用真实测量旁白时间，渲染页使用可暂停、可 seek、确定性的时间轴。禁止用整屏扫描线、长条扫光或无意义粒子代替知识动作。
+
+## 预设与扩展
+
+`general-mechanism` 是通用默认预设，而且故意不锁定视觉模板。`ai-principle-series` 只是一个经过真实系列验证的示例预设，AI 不是包的产品身份。
+
+扩展 API v1 支持 `visual`、`voice`、`research`、`qc`、`publishing`。扩展是声明式 JSON/文档/可授权资产，包含权限清单和哈希锁，不允许任意 `hooks`、`scripts` 或 `postinstall`。
+
+内置扩展：
+
+- 三个同名视觉 profile；
+- `neutral-technical-zh` 中文技术讲解声音规则；
+- `primary-source-research` 一手资料证据规则；
+- `strict-release-qc` 自动与人工发布门禁；
+- `douyin-release` 抖音标题、封面与发布资料规则。
+
+## 示例与验证
+
+包内提供两个非 AI fixture：
+
+- `credit-card-clearing`：Spatial Chamber；
+- `quantum-tunneling`：Ink Explainer。
+
+```powershell
+npm test
+npm run examples:verify
+python -m unittest discover -s python_tests -v
+$env:PYTHONUTF8='1'
+python skill\creating-explainer-videos\scripts\test_skill.py
+python skill\creating-explainer-videos\scripts\test_extensions.py
+npm run pack:local
+npm run pack:zip
+npm run smoke:packed -- .\dist\creating-explainer-videos-skill-2.0.0.tgz
+python -m build --sdist --wheel --outdir .\dist\pypi
+python -m twine check .\dist\pypi\*
+python .\scripts\audit_python_dist.py .\dist\pypi\*
+python .\scripts\smoke_pypi_package.py .\dist\pypi\creating_explainer_videos_skill-2.0.0-py3-none-any.whl
+```
+
+CI 覆盖 Windows/Linux、Node.js 18/22 和 Python 分发。发布前会分别从实际 npm `.tgz` 与 Python wheel 隔离安装，验证主/旧命令、项目脚手架、模板集、Skill 安装/校验/升级/回滚/卸载和全局命令。
+
+## 许可证与安全边界
+
+原创代码和文档采用 [MIT License](LICENSE)。字体、GSAP 商业插件、音乐、音效、图片、视频和供应商凭据不在包内。付费或异步生成必须显式授权、保存任务 ID、先查询既有任务再决定是否重试。
+
+详见 [安全策略](SECURITY.md)、[贡献指南](CONTRIBUTING.md)、[更新记录](CHANGELOG.md) 和 [v2.0.0 Release Notes](docs/releases/v2.0.0.md)。

creating_explainer_videos_skill-2.0.0/README.en.md

@@ -0,0 +1,62 @@
+# Explainer Video Skill
+
+An installable, extensible, executable Agent Skill and project scaffold for mechanism, process, system, science, and technical explainer videos.
+
+This is not a Markdown-only prompt pack. It ships a production state machine, JSON-first project generator, narration normalization, measured-audio timing import, three programmable visual templates, deterministic HTML/SVG/GSAP rendering, HyperFrames/FFmpeg adapters, media QC, declarative extensions, release gates, and non-AI end-to-end fixtures.
+
+## Install
+
+Recommended isolated CLI installation:
+
+```powershell
+pipx install creating-explainer-videos-skill==2.0.0
+explainer-video-skill --version
+```
+
+Or install into the current Python environment:
+
+```powershell
+python -m pip install creating-explainer-videos-skill==2.0.0
+```
+
+Python lifecycle commands work without Node.js. Project creation, template preview, rendering, and media commands execute the bundled canonical JavaScript runtime and require Node.js 18+.
+
+The verified npm tarball is also available from the GitHub Release:
+
+```powershell
+npm install --global https://github.com/wzwailr/creating-explainer-videos-skill/releases/download/v2.0.0/creating-explainer-videos-skill-2.0.0.tgz
+```
+
+Install the Skill into Codex or another folder-based Agent host:
+
+```powershell
+explainer-video-skill install --target codex --json
+explainer-video-skill install --destination "C:\path\to\skills" --json
+```
+
+## Create a project
+
+```powershell
+explainer-video-skill doctor --json
+explainer-video-skill new ".\demo" --title "Credit-card clearing" --topic "Why authorization is not settlement" --template spatial-chamber --preset general-mechanism --json
+explainer-video-skill status --json ".\demo"
+explainer-video-skill next --json ".\demo"
+```
+
+Agents advance through `status -> next -> real work -> evidence -> validate`. The production stages cover brief, evidence, mechanism mapping, canonical narration/cues, measured audio timing, scene design, renderer, media render, automated QC, human listening, publishing package, and explicit human release decision.
+
+Automation cannot approve a release. It can only produce `release_candidate_pending_human_listen`; `passed` requires a named human who reviewed the exact artifact hash.
+
+## Visual template collection
+
+- `paper-theatre`: object identity, evidence, sorting, comparison, and physical processing.
+- `spatial-chamber`: routes, layers, pipelines, networks, queues, and spatial state transitions.
+- `ink-explainer`: derivations, formulas, causal relations, comparison, annotation, and correction.
+
+The templates have distinct DOM and motion fingerprints, cover grammars, plugin fallbacks, and measurable QC rules. They are not lettered recolors.
+
+## Portable boundaries
+
+The package does not redistribute fonts, premium GSAP plugins, music, sound effects, stock media, or credentials. DingTalk JinBuTi/DingTalk Sans and Noto Sans SC may be used when legally installed on the host. Paid or asynchronous providers require explicit authorization and task-ID recovery rules.
+
+See the [Chinese README](README.md), [visual collection](skill/creating-explainer-videos/references/visual-template-collection.md), [extension API](skill/creating-explainer-videos/references/extension-api.md), [v2 migration](docs/MIGRATION_V2.md), and [release notes](docs/releases/v2.0.0.md).

creating_explainer_videos_skill-2.0.0/README.md

@@ -0,0 +1,158 @@
+# Explainer Video Skill
+
+[![CI](https://github.com/wzwailr/creating-explainer-videos-skill/actions/workflows/ci.yml/badge.svg)](https://github.com/wzwailr/creating-explainer-videos-skill/actions/workflows/ci.yml)
+[![PyPI](https://img.shields.io/pypi/v/creating-explainer-videos-skill)](https://pypi.org/project/creating-explainer-videos-skill/)
+[![Release](https://img.shields.io/github/v/release/wzwailr/creating-explainer-videos-skill)](https://github.com/wzwailr/creating-explainer-videos-skill/releases)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
+
+一套通用、可安装、可扩展、真正可执行的解释型视频 Agent Skill 与工程脚手架。它适用于机制、流程、系统、科学和技术科普，不绑定某个系列或某类题材。
+
+它不只是几份 Markdown：包内包含生产状态机、项目生成器、旁白规范化、真实音频时轴导入、三套可编程视觉模板、确定性 HTML/SVG/GSAP 渲染器、HyperFrames/FFmpeg 适配、媒体质检、扩展 API、打包工具和非 AI 端到端示例。
+
+[English](README.en.md) · [v2 迁移指南](docs/MIGRATION_V2.md) · [视觉模板集](skill/creating-explainer-videos/references/visual-template-collection.md) · [扩展 API](skill/creating-explainer-videos/references/extension-api.md)
+
+## 一键使用
+
+推荐使用 `pipx` 安装独立 CLI：
+
+```powershell
+pipx install creating-explainer-videos-skill==2.0.0
+explainer-video-skill --version
+```
+
+也可以安装到当前 Python 环境：
+
+```powershell
+python -m pip install creating-explainer-videos-skill==2.0.0
+explainer-video-skill --version
+```
+
+Python 入口可在没有 Node.js 时完成 Skill 的安装、更新、校验、回滚、卸载和扩展列表。创建视频工程、模板预览、渲染和媒体命令使用包内同一套 JavaScript runtime，需要 Node.js 18+。
+
+需要 npm 形式时，可直接安装 GitHub Release 中经过验证的 tarball：
+
+```powershell
+npm install --global https://github.com/wzwailr/creating-explainer-videos-skill/releases/download/v2.0.0/creating-explainer-videos-skill-2.0.0.tgz
+```
+
+安装到 Codex：
+
+```powershell
+explainer-video-skill install --target codex --json
+explainer-video-skill verify --target codex --json
+```
+
+安装到任何支持目录型 `SKILL.md` 的 Agent、仓库或团队目录：
+
+```powershell
+explainer-video-skill install --destination "C:\path\to\skills" --json
+explainer-video-skill verify --destination "C:\path\to\skills" --json
+```
+
+`--destination` 指向 skills 父目录，实际安装目录是 `creating-explainer-videos`。升级、回滚和卸载均保留可恢复备份。
+
+## 直接创建视频工程
+
+```powershell
+explainer-video-skill doctor --json
+explainer-video-skill templates list --json
+explainer-video-skill new ".\my-video" `
+  --title "信用卡清算" `
+  --topic "为什么授权成功后钱还没有到账" `
+  --template spatial-chamber `
+  --preset general-mechanism `
+  --json
+explainer-video-skill status --json ".\my-video"
+explainer-video-skill next --json ".\my-video"
+```
+
+工程会生成 JSON-first 生产合同、状态机、可运行 renderer、独立 cover、模板资产、旁白/cue 文件、媒体/QC/发布目录。Agent 按以下闭环推进：
+
+```text
+status -> next -> 执行真实工作 -> 写入证据 -> validate -> 下一阶段
+```
+
+核心阶段：
+
+```text
+brief -> evidence -> mechanism_map -> narration_and_cues
+-> real_audio_timing -> scene_spec -> runnable_renderer
+-> render -> automated_qc -> human_listen
+-> publishing_package -> human_release_decision
+```
+
+自动化只能生成 `release_candidate_pending_human_listen`。只有人工完整听看过同一个 SHA-256 成片并明确记录决定，才能成为 `passed`。
+
+## 视觉模板集
+
+| 模板 | 最适合的知识结构 | 核心动作 |
+| --- | --- | --- |
+| `paper-theatre` | 对象身份、证据、分拣、比较、规则、物理处理 | 剪、折、盖章、堆叠、交接 |
+| `spatial-chamber` | 路由、层级、管线、队列、网络、状态迁移 | 镜头纵深、路径飞行、分层变换 |
+| `ink-explainer` | 推导、公式、因果关系、比较、纠错、密集讲解 | 手绘、连接、圈注、推导、修正 |
+
+三者是结构和运动语法不同的模板，不是 A/B/C 换色版。每套都包含独立 DOM 指纹、场景 CSS、封面 CSS、时间轴控制器、插件降级和可量化 QC 规则。Spatial Chamber 正式收录了原 B 方案的纵深、路径和空间叙事能力。
+
+钉钉进步体/DingTalk Sans 可作为本机显示字体，Noto Sans SC 可作为字幕字体；仓库不分发字体文件。GSAP 商业插件同样不随包分发，每个模板都声明了开源/原生降级路径。
+
+## 旁白、动画与渲染
+
+字幕与 TTS 来自同一个 canonical 字符串。规范化会移除 `\_`、snake_case 下划线和可能被读成符号名的 Markdown 标记。真实旁白生成后导入测量时轴：
+
+```powershell
+explainer-video-skill narration prepare ".\my-video" --json
+explainer-video-skill narration import-timing ".\my-video" --timing ".\timing.json" --json
+explainer-video-skill build ".\my-video" --json
+explainer-video-skill render ".\my-video" --json
+explainer-video-skill cover ".\my-video" --json
+explainer-video-skill mux ".\my-video" --audio ".\narration.wav" --json
+explainer-video-skill audit ".\my-video" --json
+explainer-video-skill package ".\my-video" --json
+```
+
+最终动画只使用真实测量旁白时间，渲染页使用可暂停、可 seek、确定性的时间轴。禁止用整屏扫描线、长条扫光或无意义粒子代替知识动作。
+
+## 预设与扩展
+
+`general-mechanism` 是通用默认预设，而且故意不锁定视觉模板。`ai-principle-series` 只是一个经过真实系列验证的示例预设，AI 不是包的产品身份。
+
+扩展 API v1 支持 `visual`、`voice`、`research`、`qc`、`publishing`。扩展是声明式 JSON/文档/可授权资产，包含权限清单和哈希锁，不允许任意 `hooks`、`scripts` 或 `postinstall`。
+
+内置扩展：
+
+- 三个同名视觉 profile；
+- `neutral-technical-zh` 中文技术讲解声音规则；
+- `primary-source-research` 一手资料证据规则；
+- `strict-release-qc` 自动与人工发布门禁；
+- `douyin-release` 抖音标题、封面与发布资料规则。
+
+## 示例与验证
+
+包内提供两个非 AI fixture：
+
+- `credit-card-clearing`：Spatial Chamber；
+- `quantum-tunneling`：Ink Explainer。
+
+```powershell
+npm test
+npm run examples:verify
+python -m unittest discover -s python_tests -v
+$env:PYTHONUTF8='1'
+python skill\creating-explainer-videos\scripts\test_skill.py
+python skill\creating-explainer-videos\scripts\test_extensions.py
+npm run pack:local
+npm run pack:zip
+npm run smoke:packed -- .\dist\creating-explainer-videos-skill-2.0.0.tgz
+python -m build --sdist --wheel --outdir .\dist\pypi
+python -m twine check .\dist\pypi\*
+python .\scripts\audit_python_dist.py .\dist\pypi\*
+python .\scripts\smoke_pypi_package.py .\dist\pypi\creating_explainer_videos_skill-2.0.0-py3-none-any.whl
+```
+
+CI 覆盖 Windows/Linux、Node.js 18/22 和 Python 分发。发布前会分别从实际 npm `.tgz` 与 Python wheel 隔离安装，验证主/旧命令、项目脚手架、模板集、Skill 安装/校验/升级/回滚/卸载和全局命令。
+
+## 许可证与安全边界
+
+原创代码和文档采用 [MIT License](LICENSE)。字体、GSAP 商业插件、音乐、音效、图片、视频和供应商凭据不在包内。付费或异步生成必须显式授权、保存任务 ID、先查询既有任务再决定是否重试。
+
+详见 [安全策略](SECURITY.md)、[贡献指南](CONTRIBUTING.md)、[更新记录](CHANGELOG.md) 和 [v2.0.0 Release Notes](docs/releases/v2.0.0.md)。

creating_explainer_videos_skill-2.0.0/SECURITY.md

@@ -0,0 +1,19 @@
+# Security policy
+
+## Supported version
+
+Security fixes target the latest major release. v1 is retained as a legacy release; upgrade to v2 for current fixes.
+
+## Reporting
+
+Do not post credentials, tokens, private media, paid-provider task IDs, or personal filesystem paths in a public issue. Create a minimal non-sensitive report and request a private follow-up when sensitive evidence is unavoidable.
+
+## Trust boundaries
+
+- Extension API v1 is declarative, permission-listed, path-contained, and rejects executable hooks.
+- Skill installation verifies a SHA-256 package manifest before copying files.
+- Update, rollback, and uninstall target one exact Skill directory and preserve recoverable backups.
+- CLI process execution uses argument arrays; it does not build shell commands from project text.
+- Provider credentials and paid-execution authorization remain outside the package and generated projects.
+- Automated QC cannot create a human `passed` decision.
+- Published archives must exclude tokens, personal absolute paths, cache files, generated media, font binaries, and premium plugin binaries.

creating_explainer_videos_skill-2.0.0/docs/superpowers/plans/2026-08-31-pypi-distribution.md

@@ -0,0 +1,106 @@
+# PyPI Distribution Implementation Plan
+
+> **For agentic workers:** REQUIRED SUB-SKILL: Use superpowers:subagent-driven-development (recommended) or superpowers:executing-plans to implement this plan task-by-task. Steps use checkbox (`- [ ]`) syntax for tracking.
+
+**Goal:** Add, validate, and publish a Python distribution that installs the complete explainer-video Skill and exposes the existing command surface.
+
+**Architecture:** A dependency-free Python installer owns filesystem-safe Skill lifecycle commands. Project and render commands execute the canonical bundled JavaScript runtime after checking Node.js 18+, while Hatchling includes the canonical Skill tree directly in the wheel so assets have one source of truth.
+
+**Tech Stack:** Python 3.9+, standard library, Hatchling, unittest, Node.js 18+, Twine
+
+**Spec:** `docs/superpowers/specs/2026-08-31-pypi-distribution-design.md`
+
+## Global Constraints
+
+- Package and command identity remain `creating-explainer-videos-skill` and `explainer-video-skill` at version `2.0.0`.
+- The Python distribution contains the canonical Skill, template collection, extensions, presets, runtime, scripts, and manifest.
+- Python lifecycle commands work without Node.js; project and rendering commands require Node.js 18+.
+- Publication credentials must never be printed, copied into the repository, or included in an archive.
+- Existing npm behavior and its package file allowlist remain unchanged.
+
+---
+
+### Task 1: Python CLI and packaged-resource contract
+
+**Files:**
+- Create: `python_tests/test_cli.py`
+- Create: `python_src/creating_explainer_videos_skill/__init__.py`
+- Create: `python_src/creating_explainer_videos_skill/__main__.py`
+- Create: `python_src/creating_explainer_videos_skill/cli.py`
+- Create: `python_src/creating_explainer_videos_skill/resources.py`
+
+**Interfaces:**
+- Produces: `main(argv: Sequence[str] | None) -> int`, `package_paths() -> PackagePaths`, and the two console entrypoints.
+- Consumes: canonical Skill files from the repository in source mode and bundled `assets/` in an installed wheel.
+
+- [x] **Step 1: Write failing CLI tests**
+
+```python
+def test_version_works_without_node(self):
+    result = run_cli("--version", env={"PATH": ""})
+    self.assertEqual(result.returncode, 0)
+    self.assertEqual(result.stdout.strip(), "2.0.0")
+
+def test_project_command_reports_node_18_requirement(self):
+    result = run_cli("templates", "list", env={"PATH": ""})
+    self.assertNotEqual(result.returncode, 0)
+    self.assertIn("Node.js 18+", result.stderr)
+```
+
+- [x] **Step 2: Run `py -m unittest python_tests.test_cli -v` and confirm failure because the package does not exist.**
+- [x] **Step 3: Implement version/help handling, packaged-resource discovery, and Node command delegation.**
+- [x] **Step 4: Re-run the focused test and confirm it passes.**
+
+### Task 2: Safe Python Skill lifecycle
+
+**Files:**
+- Create: `python_tests/test_installer.py`
+- Create: `python_src/creating_explainer_videos_skill/installer.py`
+- Create: `python_src/creating_explainer_videos_skill/integrity.py`
+- Modify: `python_src/creating_explainer_videos_skill/cli.py`
+
+**Interfaces:**
+- Produces: `install_skill`, `verify_installed_skill`, `rollback_skill`, `uninstall_skill`, and `discover_extensions`.
+- Consumes: `PackagePaths.skill_root` and `PackagePaths.manifest_path` from Task 1.
+
+- [x] **Step 1: Write failing integration tests for install, update backup, manifest tampering, rollback, uninstall recovery, custom destinations, and extension traversal.**
+- [x] **Step 2: Run `py -m unittest python_tests.test_installer -v` and confirm failures name missing lifecycle behavior.**
+- [x] **Step 3: Implement staged copies, manifest verification, extension validation, recoverable backups, and JSON output.**
+- [x] **Step 4: Re-run both Python test modules and confirm they pass.**
+
+### Task 3: Build metadata and archive smoke test
+
+**Files:**
+- Create: `pyproject.toml`
+- Create: `scripts/smoke_pypi_package.py`
+- Create: `scripts/audit_python_dist.py`
+- Create: `python_tests/__init__.py`
+- Modify: `.gitignore`
+
+**Interfaces:**
+- Produces: `dist/*.whl`, `dist/*.tar.gz`, and deterministic archive audit/smoke commands.
+- Consumes: Python package and canonical Skill tree from Tasks 1-2.
+
+- [x] **Step 1: Write the smoke script expectations for wheel installation, both CLI names, templates, Skill installation, and verification.**
+- [x] **Step 2: Run the build before metadata exists and confirm it fails for the expected missing configuration.**
+- [x] **Step 3: Add Hatchling metadata, console scripts, force-included assets, and archive auditing.**
+- [x] **Step 4: Build wheel/sdist in an isolated environment, run `twine check`, audit both archives, and run the exact-wheel smoke test.**
+
+### Task 4: Documentation, release evidence, and publication
+
+**Files:**
+- Modify: `README.md`
+- Modify: `README.en.md`
+- Modify: `CHANGELOG.md`
+- Modify: `docs/releases/v2.0.0.md`
+- Modify: `docs/superpowers/plans/2026-08-31-pypi-distribution.md`
+
+**Interfaces:**
+- Produces: verified install instructions and public registry evidence.
+- Consumes: exact built archives and local PyPI credential.
+
+- [x] **Step 1: Document `pipx`/`pip` installation, Node.js boundary, dual-registry identity, and public verification commands.**
+- [x] **Step 2: Run all Python tests, existing Node tests, Skill validators, build, `twine check`, archive audit, and clean-wheel smoke test.**
+- [ ] **Step 3: Upload only the verified wheel and source distribution to production PyPI without exposing the token.**
+- [ ] **Step 4: Install `creating-explainer-videos-skill==2.0.0` from public PyPI in a fresh environment and repeat the smoke checks.**
+- [ ] **Step 5: Commit and push the verified source changes, then update the GitHub repository homepage/release links if publication succeeds.**

creating_explainer_videos_skill-2.0.0/docs/superpowers/specs/2026-08-31-pypi-distribution-design.md

@@ -0,0 +1,36 @@
+# PyPI Distribution Design
+
+## Goal
+
+Publish `creating-explainer-videos-skill` as a real Python distribution without weakening the existing npm package or duplicating the maintained Skill and template assets.
+
+## User Experience
+
+- Install with `pipx install creating-explainer-videos-skill` or `python -m pip install creating-explainer-videos-skill`.
+- Run `explainer-video-skill --help` and `explainer-video-skill --version` without Node.js.
+- Install, update, verify, roll back, uninstall, and list extensions through the Python CLI.
+- Keep `ai-principle-video-skill` as a compatibility command.
+- Delegate project creation, validation, rendering, and media commands to the bundled JavaScript runtime, with a clear Node.js 18+ diagnostic when Node is unavailable.
+
+## Architecture
+
+The Python package is an additional distribution layer in the same repository. Hatchling builds a pure-Python wheel from `python_src/creating_explainer_videos_skill` and force-includes the canonical `skill/creating-explainer-videos` tree plus `skill-manifest.json` as package assets. This avoids a second maintained copy of the Skill, templates, extensions, and runtime.
+
+The Python installer mirrors the safety properties of the Node installer: validate before mutation, stage before replace, back up an existing installation, keep uninstall recoverable, reject path traversal in extension entrypoints, and verify every packaged file against the canonical manifest.
+
+## Boundaries
+
+- Python 3.9+ is required for the Python distribution.
+- Node.js 18+ remains required for project and rendering commands because the canonical runtime, GSAP templates, and rendering toolchain are JavaScript-based.
+- No credentials, `.npmrc`, `.pypirc`, fonts, media binaries, caches, or machine-specific paths may enter wheel or source distribution files.
+- The npm tarball allowlist remains unchanged.
+- PyPI publication uses the local credential only at upload time; it is never printed, committed, or copied into project files.
+
+## Verification Gates
+
+1. Python unit and integration tests pass.
+2. Existing Node tests and Skill validators pass.
+3. Wheel and sdist build successfully and pass `twine check`.
+4. A clean virtual environment installs the exact wheel and exercises version, help, template listing, Skill installation, and verification.
+5. The built archives pass a secret and portability audit.
+6. After upload, a second clean environment installs the exact public PyPI version and repeats the smoke checks.