Edit and render video from ChatGPT
Add Rendobar to ChatGPT on the web as a custom MCP server, and ChatGPT runs FFmpeg commands and video renders from the chat. Sign-in is OAuth, so there is no key to paste.
Overview
Describe the edit and ChatGPT writes the job, Rendobar runs it, and the chat gets a link to the finished file. That covers FFmpeg commands, timeline videos, captions and compression. The MCP server docs list every tool.
Set up
Add the server
In ChatGPT on the web, open Plugins, press the plus button, choose Add custom MCP server and paste the server URL above.
Pick OAuth
Set authentication to OAuth, accept the risk warning and select Create as a plugin.
Install and ask
Install the plugin, approve Rendobar in your browser when asked, then describe the edit in a chat.
Questions
Is Rendobar a listed ChatGPT plugin?
Not yet. You add it yourself as a custom MCP server with the address https://api.rendobar.com/mcp, from ChatGPT on the web.
Will ChatGPT ask before it runs a job?
Yes, by default. Rendobar marks each tool that submits, probes or cancels a job, or opens or finishes an upload, as a write action, so ChatGPT asks you to confirm each call. Reads such as job status and your balance run without asking. A remembered approval lasts for that conversation only.
Can ChatGPT use a video that is not online yet?
Yes. ChatGPT starts an upload and sends you a link to a page where you pick the file on any device, your phone included. The job then runs on that file.
Related
- Claude Ask Claude to cut a video, add captions or make an image, and get the file back in the chat.
- Cursor Let your coding agent run real media jobs while it builds the feature.
- n8n Run any Rendobar job in a workflow, from FFmpeg to captions and AI images.
