Video Skills for Codex · Claude Code · Cursor · WorkBuddy

Teach your agent
to make video

One sentence installs it into the AI assistant you already use. Research, script, voiceover, final cut: all on your own machine. Skills are free, editable, and shareable.

See the 7 official Skills
  • Skills and the local app are free
  • Renders on your machine
  • Credits only for voice and generation
Codex · local taskLocal

Use this ad as the reference and make a 45-second, 9:16 UGC video for my serum.

Using $ecommerce-ugc-remake-advanced. Two quick questions: which licensed talent assets go on camera? Any claims that must be made or avoided?

Pick a video like the one you need, then see how it was made

Real outputs. Hover to play, click for the process.

1 storyboard frame → 1 music track → rendered locally · author noted 5 steps

一张产品图加三个卖点,做成 20 秒卡点动效广告

made by 产品动效片 · Product Motion Reel

21s·9 credits

2 video shots → 1 music track → voice-over → speech to text → align subtitles → rendered locally · you confirmed once

徕芬 SE 高速负离子吹风机真实种草口播视频

made by AI 3C 种草

44s·41 credits

2 storyboard frames → voice-over → align subtitles → rendered locally · author noted 4 steps · you confirmed 3 times

给社区 Skill 再做条推广片:直播间腔调的口播讲购买力挑战,内录制作记录,结尾接成片

made by Skill Promo Video

90s·8 credits

2 storyboard frames → 2 video shots → rendered locally · you confirmed once

一张商品图,复刻爆款参考片

made by 视频复刻高级版

10s·34 credits

2 storyboard frames → voice-over → align subtitles → rendered locally · author noted 11 steps · you confirmed once

给社区 Skill 做条推广片:内录制作记录、名人音色口播,结尾接成片

made by Skill Promo Video

81s·12 credits

1 storyboard frame → 2 video shots → rendered locally · author noted 1 step

一张商品图,复刻爆款参考片

made by 视频复刻基础版

15s·65 credits

speech to text → rendered locally · author noted 1 step

口播和录屏自动对齐,剪成专业解说视频;三角洲实例

made by Presenter + Screen Sync Video

83s·9 credits

align subtitles → rendered locally · author noted 1 step

复用确认素材与配音,核对字幕对齐并渲染营销号视频

made by Reference Marketing Video Pro — 0.3.0

16s·1 credits

voice-over → align subtitles → rendered locally · author noted 1 step · you confirmed once

用真实产品内录讲清营销视频制作操作,配合小林说旁白、重点框选和平滑居中

made by Product Screen Video

72s·4 credits

speech to text → rendered locally

适用于人物访谈、视频播客和专家对谈,将长视频剪成聚焦说话人的竖屏短片,并用字幕突出核心观点。

made by Interview Shorts

60s·2 credits

analyze the reference → 1 storyboard frame → rendered locally · author noted 4 steps

把自己的深蹲和扣篮素材剪成竖版高燃集锦,扣篮顶点做子弹时间慢动作

made by Sports Hype Reel

24s·5 credits

1 music track → rendered locally · you confirmed once

只需录屏素材就能直接出片的种草skill

made by 技能安利剪辑

21s·6 credits

Three steps from a sentence to a finished video

  1. Copy one sentence, paste it into your agent

    Your agent downloads the build, verifies the signature, and installs every official Skill in one go.

  2. Say what video you want

    Just say it the way you would to a colleague. The entry Skill picks the right method and checks in with you all the way to the final cut.

  3. The video renders on your computer

    Projects, timelines and rendering stay on your machine; only voice, image and video generation are billed in credits, quoted before each step. The first time one is used you click Authorize once in your browser — nothing to manage after that.

Tools: every capability your Agent can use

What exists, what it costs, what to say to Codex.

Video3

AI avatarScaleMusefrom 1 credit per second
Try it

AI avatar

Video

One front-facing photo (or an existing video) plus an audio clip — a lip-synced talking video

Multi-asset videoScaleMusefrom 1 credit per second
Try it

Multi-asset video

Video

Up to 9 images and 3 audio clips plus one sentence — a 1–15 second video

Motion transferScaleMuse1 credit per second, up to 30 s
Try it

Motion transfer

Video

One portrait photo plus a dance or action video — the motion moves onto the person in the photo

Audio3

Voice-overScaleMusefrom 1 credit per minute (premium voices 7)
Try it

Voice-over

Audio

Read text aloud in dozens of voices

Speech to subtitlesScaleMuse1 credit per minute
Try it

Speech to subtitles

Audio

Turn the speech in a video or audio file into timed text

Subtitle alignmentScaleMuse1 credit per minute
Try it

Subtitle alignment

Audio

Align a script to the audio line by line

Video tools & assets5

Erase subtitles & watermarksScaleMusefrom 4 credits per minute
Try it

Erase subtitles & watermarks

Video tools

Remove subtitles, watermarks and logos from a video

Multilingual dubbingScaleMusefrom 3 credits per language per minute
Try it

Multilingual dubbing

Video tools

Translate one video into many languages: transcribe, translate, dub, erase and re-burn subtitles — one finished video per language

On-screen text translationScaleMusefrom 11 credits per minute
Try it

On-screen text translation

Video tools

Translate only the text in the frame (titles, stickers, packaging); the voice stays

Break down a hit videoScaleMuse1 credit per second, 15 max
Try it

Break down a hit video

Video tools

Split a hit into shots, camera moves, second-by-second action and ready-to-use prompts

Upload assetsScaleMusefree
Try it

Upload assets

Assets

Hand files on your computer to the Agent; cleared after 24 hours

Most of it is free. The paid part is spelled out.

  • Skills and the local app: free
  • Local projects, timelines, and rendering: free
  • Voiceover, speech recognition, image and video generation: credits, quoted before you submit

Frequently asked questions

Do I need an account?

Not to install Skills or the local program. An account is needed only when the Agent uses voice, image or video generation: it hands you a link, you log in and click Authorize once — no key to create or paste.

Which agents and systems are supported?

Today: Codex, Claude Code, Cursor, and WorkBuddy on macOS (Apple Silicon) and Windows. Paste the same install prompt into whichever agent you use. Skills install to ~/.agents/skills, which Codex and Cursor read directly; Claude Code reads the same copy through ~/.claude/skills, WorkBuddy through ~/.workbuddy/skills. Linux is still on the roadmap.

Do my files get uploaded?

No. Projects, timelines, and renders stay on your computer. Only the model capabilities you approve send the necessary inputs to ScaleMuse, and you see the estimated credits first.

Can I still make videos in the browser?

Yes. The web workspace (one-line creation, video remake, multilingual translation, smart erase) works as before and shares one account and credit pool with the Skills.

FE 6d45f563-20260930071721 | BE ...