OpenChatCut
OpenChatCut is an open-source, local-first AI video editor that uses conversational agents (like Codex or Claude) to edit a real multitrack timeline. You describe the cut, and the agent places clips, trims, adds captions, and builds transitions—every step remains editable and reversible.
Open-source AI agent video editor with a real timeline
- Quickly assemble rough cuts from raw footage by describing the desired pace and style.
- Automate repetitive editing tasks like adding captions, music, and transitions.
- Collaborate with AI agents to refine timeline edits without losing manual control.
- Create polished videos for social media, presentations, or internal projects.
- Experiment with different AI models (Codex, Claude) to compare editing workflows.
- Download the desktop build or clone the repository from GitHub.
- Launch OpenChatCut and open or create a project with your media files.
- Type or speak your editing request (e.g., 'Add captions, trim dead space, and lay in background music').
- The agent reads the project state and applies changes directly to the timeline.
- Review the edits, undo any step, or refine details manually using familiar editing controls.
- Export the final video as an editable project or flattened file.

Full-Stack Platform for Training Small Language Models


Vibe code business operations


Your whole stack, running itself!


The easiest way to build AI agents for all your tasks


Your people, your agents, your project — all in one place


Email for AI agents. They sign up, send and reply.


Turn any webpage into an action manifest for AI agents


One-stop live data marketplace for your agent


A markdown editor you and your agent co-edit


Instant, hardware Isolated Sandboxes for AI agents


The AI agent that handles checkout.


Your personal context file for every agent.


Build AI agents, workflows, and apps in one stack


Launch iMessage agents in seconds


An AI coworker with its own cloud computer


Your AI agent for autonomous web app testing


Annotate anything for humans and their agents


Scale your creativity on editable canvas with agent memory





