Video Use lets you edit videos with coding agents. From the team behind Browser Use, it brings the agent workflow to video production: instead of dragging clips on a timeline for hours, you describe the edit and let an AI coding agent assemble, trim and refine the video for you.
Video Use has attracted more than 29,400 stars on GitHub in a short time and is actively developed, with updates in October 2026. The idea follows a simple observation: coding agents are already good at manipulating files, following instructions and iterating on feedback, and video editing is largely file manipulation with creative judgment layered on top. Video Use gives agents the tooling to work with video footage directly, cutting clips, arranging sequences, applying edits and re-rendering in response to plain-language feedback. Because the agent works through code and commands, edits are reproducible and easy to adjust: change the instruction, re-run, and compare, rather than undoing timeline operations by hand.
Video Use is free and open source under the MIT license. Your only costs are the coding agent or model subscription you already use to drive it.
Video Use is for creators, developers and social media teams who produce video regularly and want agent-speed rough cuts and batch edits, and for tinkerers exploring what happens when coding agents get media tools.