Documentation

OpenVideo documentation

Install OpenVideo from source, learn the workspace and timeline, put the Edit Agent to work, and connect only the model providers you want.

OpenVideo runs from source today — there is no packaged installer yet. Everything documented here is behaviour this build actually has; planned work is called out as not yet available.

Start

What OpenVideo is: A local-first desktop video editor with an agent that can operate the timeline instead of only advising you.
Install and run: Prerequisites, cloning the repository, pointing OpenVideo at your FFmpeg, and verifying the build from source.
Projects and folders: A project is a folder you choose. Media, chat history, and generated results are written inside it.

Editing

The workspace: A tour of the editing surface: media dock, Program Monitor, inspector, timeline, and the docked agent chat.
Timeline editing: Importing media, arranging clips, keyframes and transitions, and the default keyboard shortcuts.
Exporting: How OpenVideo renders an MP4 through your own FFmpeg, and what it refuses to start.

Edit Agent

The Edit Agent: A chat panel that operates the editor through a typed tool surface, and asks approval before it writes.
Agent tools: The typed operations the Edit Agent can call, and which of them ask for approval first.

Models and generation

Providers and models: Connect a local engine, an API key, or a ChatGPT sign-in — and see only the models you actually have.
Voice generation: Write a script, pick a voice model, and import the result straight into the open project.
Video generation: Prompt with a style, aspect ratio, and duration — optionally seeded by a reference image.

Reference

Settings: Every settings section, from theme and local tool readiness to providers, shortcuts, and updates.
Data and privacy: Where your footage lives, what crosses the process boundary, and what is never collected.
Troubleshooting: Export failures, missing FFmpeg, capture permissions, provider errors, and agent limits.