Claude Code builds a local video compression interface
Willison wanted to optimize a phone recording before publishing it. He used Claude Code for web to build a browser tool around FFmpeg compiled to WebAssembly, with multiple output presets.
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Build a local video compression page using [approved FFmpeg build]. Offer a few clearly labeled quality presets, output sizes and previews. Test with a non-sensitive sample, show the actual command for each output and report unsupported browsers. Keep processing local; do not upload the input.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source author
- Simon Willison
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- The source shows a tool producing several encoded versions with previews and downloads; no general speed claim is made here.
- Human involvement
- Willison chose the recording and evaluated output options. Publication approval boundaries are not described.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- Use the need to compress a specific phone recording as the implementation goal.
- Build a browser interface around the WebAssembly version of FFmpeg.
- Generate different encodings and compare previews and file sizes to choose a usable output.
More details
About this account
The coding agent built the tool from the delegated goal; the short account links the session but does not detail its test loop.
Firsthand tool note and screenshot document use for a real recording. Quality and timing shown are specific to that input.
Credit: Simon Willison
Source published
Added to Agents In Use
Source checked
Tools and topics
Claude Code · FFmpeg · WebAssembly
Browser · Media
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.