---

name: lov-video-chapter-export description: Package chapter overlays, timestamps, CSV metadata, and project JSON for 剪映/CapCut and other video editors. Use when the user asks for editor integration, transferable chapter assets, or a ready-to-import chapter package. depends_on:

• lov-branding-consistency

---

章节打包 · Chapter Export

Run:

python3 ../../scripts/render_chapter_bar.py package \
  --project "/path/to/chapter-project.json" \
  --output "/path/to/chapter-package"

For 剪映/CapCut:

1. Import chapter-overlay.mov.
2. Place it on the top video track at 00:00.
3. Keep scaling at 100% and match the project frame rate.
4. Trim only the tail when the editor rounds duration.

Use chapters.txt for platform chapter fields and chapters.csv for editor automation. Treat direct draft-file writing as an experimental adapter because editor project formats can change independently.

Runtime context (shared)

运行前读取本 Skill 包的 skill.yaml，由宿主提供 skill-runtime/v1 上下文。字段解析顺序为：当前请求、项目上下文、个人 Preferences、品牌 Profile、通用默认值。

• 只使用 Manifest 声明的字段；Profile 保存公开品牌事实，Preferences 保存个人工作偏好。
• required: true 字段缺失时，按 Manifest 的问题配置向用户提出一个聚焦问题；用户明确同意后再保存回答。
• 报错提供可复制的 context_id、字段路径与来源，诊断内容避开秘密、完整私人路径和原始配置。