remotion-captions Skill for Claude: What It Does and How to Install It
Makes Claude turn audio and video into captions, import .srt files, and show them in a Remotion video.
- Installs
- 251.2K
- on skills.sh
- Stars
- 6K
- on GitHub
- Files
- 5
- besides SKILL.md
- Updated
- Oct 2026
- last repo commit
remotion-captions covers three jobs: generating captions from audio or video, importing them from an .srt file, and displaying them in a Remotion video. All captions are handled as JSON in Remotion's Caption type, with text, start and end in milliseconds, a timestamp, a confidence value and an optional page break.
Transcription runs Whisper locally on your GPU through the @remotion/whisper-webgpu package, so you need a compatible GPU. The guide says it does not work on Linux arm64. Claude writes a Node.js script that decodes the audio, downloads a model and transcribes. After captions are done, SKILL.md says to load Remotion Best Practices and to open Remotion Studio, so install the whole Remotion set.
Good for
- Developers adding subtitles to Remotion videos
- Creators who have an .srt file and want it on screen
Install remotion-captions
Claude Code
- 1
Copy this prompt.
Install the remotion-captions skill in Claude Code. Run this command: npx skills add remotion-dev/skills Then check that it loaded and tell me how to use it.
- 2
Paste it into Claude Code. It shows the change and asks before running it.
Requires Node.js.
Install the whole set, not just remotion-captions. This skill hands off to other Remotion skills in the same repo, such as best practices, and the README says the set works with Claude Code, Codex, Kimi Code and Cursor.
From the project README. Source.
What's in the skill
When it loads
Transcribing, displaying and animating captions
Sections in SKILL.md
- Generating captions
- Displaying captions
- Importing captions
- Post-creation
Key facts
- Author
- remotion-dev
- Installs on skills.sh
- 251,208 (Nov 2026)
- GitHub stars
- 6,037 (Nov 2026)
- Last repo commit
- Oct 26, 2026
- Files in the skill folder
- 6 including SKILL.md
- skills.sh audits
- Gen Agent Trust Hub: pass, Socket: pass, Snyk: pass
- Needs for transcription
- A GPU that supports WebGPU Whisper; not Linux arm64
What it includes
Keeps all captions as JSON in the Remotion Caption type
Caption formatWrites a script that decodes audio, downloads a Whisper model and transcribes
Runs code ·
Generate captions (transcribe-captions.md)Parses an .srt file into the Caption format with parseSrt()
Read ·
Import .srt (import-srt-captions.md)Shows captions in the video using the Basic Captions element
Write ·
Display captions (display-captions.md)
From the official docs
Permissions
Medium risk
Claude adds npm packages, downloads a speech model and writes a script and a captions file into your project. Review what it installs before you accept.
What it can do
- Read your data
- Create and change things
- Delete things
- Read-only mode
Watch out for
It adds packages and downloads a speech model
The transcription guide installs @remotion/whisper-webgpu, @huggingface/transformers, mediabunny and @mediabunny/server, then downloads a model. Check the package list in your diff.
It needs a compatible GPU
The guide says a compatible GPU is required and ONNX Runtime does not work on Linux arm64.
Technical details
| Talks to | npm registry. Whisper model download |
|---|
Questions
How do I install remotion-captions?
Use the install box above. It installs the whole Remotion set, because SKILL.md tells Claude to load Remotion Best Practices when it finishes.
Does transcription send my audio to a cloud service?
The guide says Whisper runs locally on your GPU through @remotion/whisper-webgpu. It does download a model first.
What do I need to run it?
A compatible GPU, per the transcription guide. It also says ONNX Runtime does not work on Linux arm64. Its install step adds @remotion/whisper-webgpu, @huggingface/transformers, mediabunny and @mediabunny/server.
Can I use captions I already have?
Yes. SKILL.md links to a separate guide, import-srt-captions.md, which parses an .srt file with parseSrt() from @remotion/captions.
Is it safe?
It adds npm packages with npx remotion add, downloads a Whisper model (small.en in the guide's script) and writes a script and a captions123.json file in the guide's example. Check the package list in your diff before you accept.
Information
updated Nov 2, 2026
- Repository
- remotion-dev/skills
- Folder
skills/remotion-captions
Sources
Docs, issue trackers and discussions this page is built from.
Show all 5Hide
- remotion-dev/skills README · github.com
- remotion-captions SKILL.md · github.com
- remotion-captions transcribe-captions.md · github.com
- remotion-captions import-srt-captions.md · github.com
- remotion-captions display-captions.md · github.com