Make specific edits to existing chengfeng-videocut projects through its Runtime CLI. Locate spoken passages, add or remove visual overlays, delete specified content, split or move clips, and adjust canvas proportions, then read back the affected project data to verify changes.
操作 chengfeng-videocut 剪辑工作台的已有工程，通过 Runtime CLI 查工程与词句位置、添加或替换画面覆盖层、移除画面、删除明确指定的词句或片段、拆分裁剪移动已有片段、调整画布比例并回读核验。
用户说插入片段、删掉这句、移除动画、换个画面、调整顺序或管理时间线时使用；先区分画面覆盖与主轨顺延插入，不负责自动判断口误、创作动画、制作字幕、导出或排查安装故障。
Review the instructions the agent will receive before you install.
把明确的编辑要求变成现有 CLI 操作，始终在同一工程回读结果。不是另一套编辑引擎，也不是每次使用其他 Skill 前的必读入口。
| 用户要做什么 | 本 Skill 的路径 |
|---|---|
| 看当前工程、找一句话、看已有片段和画面 | 只读定位，见 操作方法 |
| 在一句话上放动画或视频，保留原口播 | 画面覆盖；时间线不加长，见 操作方法 |
| 删除画面、词句或片段 | 区分对象与声音影响，见 删除方法 |
| 分成两段、缩短一段、调换已有片段顺序 | EDL 修改，见 操作方法 |
| 改横竖比例、画面铺满或完整显示 | 分清工程画布和媒体 fit，见 操作方法 |
| 在主时间线任意位置插新素材并顺延 | 先查当前能力；没有通用插入命令就明确缺口，不把覆盖或移位冒充插入 |
用户说“插入”时，先判断是盖画面且口播继续，还是添加新时长、让后面顺延；说“删除”时，先判断是移除画面还是剪掉声音与时间。意图已明确就执行，不重复追问；重复句、重用素材或目标不唯一时，列出位置供确认，不凭文本猜实例。
需要自动判断口误、重复和保留哪一版，交给剪口播 Skill；字幕内容与排版交给字幕 Skill；镜头/动画创作交给相应设计 Skill；成片输出交给导出 Skill；连接故障和升级交给维护与排障。不自动启动额外 Agent，也不自动安装缺失 Skill。
workbench commands --json 的字段和模式调用。写操作的 --confirmed 表示已有授权，不是授予 Agent 权限的开关。编辑正式工程只通过 Runtime，不直接改工程 JSON。playback 是读取播放词流，operation-recover 是受限事务恢复，不是任意 undo。No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Check whether the video editing skills and their runtime are compatible and ready to use. Handle runtime installation, diagnose missing tools, guide transcription credential setup, and activate verified skill updates after user confirmation.
Prepare a reviewed cut list for Chinese talking videos by transcribing words, correcting recognition errors, and identifying repetitions or verbal mistakes. Present deletion tables for human review in Studio without cutting or exporting media.
Export an edited talking video as an MP4 with cuts, zooms, subtitles, and HTML visual layers rendered together. Preview the export plan first, then check output properties and inspect frames from the finished file.
Prepare bug reports for chengfeng-videocut by organizing reproduction steps, expected behavior, and minimal diagnostic evidence. Review a draft with sensitive information removed, then submit a GitHub Issue only after explicit confirmation, with checks for existing matching reports.
Create subtitle data for edited talking videos using an existing word transcript and edit list. Correct names, split text into readable screens, and review wording and timing in Studio without retranscribing or rendering a finished video.
Add visual explanations to edited spoken videos and screen recordings using HTML overlays tied to subtitle segments. Inspect actual frames and screen movement to place highlights or full-screen animations, then check each layer in the preview.