New: the Visuall MCP server
Make AI videos from Claude, ChatGPT or Gemini
Connect Visuall to Claude, ChatGPT or Gemini and ask for a film in the chat. Your assistant can have Visuall's AI Director make it, or build any canvas box by box: your photos in, keyframes, models compared, picks, stitching, 4K. Everything lands on your Visuall canvas, where you can change any step and run it again.
Set it up in three steps
- 1
Copy the server address.
https://api.visuall.ai/mcp - 2
In Claude, open Settings → Connectors, choose Add custom connector and paste the address.
- 3
Sign in to Visuall and choose Allow.
No API key to paste. Films use your Visuall credits, and the assistant tells you the price before it makes one. New accounts start with enough free credits for a 5-second film.
Things to ask
“Make a 15-second vertical film for Reels: a street cat in Istanbul steals a simit and gets away across the rooftops.”
“Turn this product story into a 10-second ad: our cold brew, poured over ice on a sunny balcony, slow and glossy.”
“Build a canvas from these three product photos: 4K keyframes, each shot filmed with two video models so I can compare, then stitch my picks.”
“Show me my last three Visuall films and share the best one.”
What the assistant can and can't do
It can make films, build and change canvases box by box, add your photos from links, price and run them, follow the run, re-make a single box, and make share links when you ask.
It can't buy credits, change your sign-in or see your payment details. Every film it makes is on your canvas like any other.
To disconnect an app, open Settings and choose Disconnect next to it.