What to type.
Add this song under the video, quieter when I talk, and fade it out at the end.
See it: Edit a dance Short to the beat
Give it the song. The agent lays it under the video, lowers it while someone is talking, fades it in and out, and can cut the video to its beats.
LiveWatch it live now Kazi gives Big Buck Bunny a happier ending.Kazi is one way in: ChatGPT, Claude or any MCP app connects the same way.Free · a few minutes · no sign-upRather watch first? Watch Kazi edit one live, free
On your phone? ChatGPT adds cuttingfloor on a computer. Try it now in Kazi, free
Kazi works on your phone and opens with a sample clip and edits it on cuttingfloor.
Or set it up later: Send me the setup
Rather watch first? Watch Kazi edit one live, free
On your phone? Claude adds cuttingfloor on a computer. Try it now in Kazi, free
Kazi works on your phone and opens with a sample clip and edits it on cuttingfloor.
Or set it up later: Send me the setup
Add this song under the video, quieter when I talk, and fade it out at the end.
See it: Edit a dance Short to the beat
https://api.cuttingfloor.ai/mcpchatgpt.com/plugins → + → Add custom MCP server. Paste the address, set Authentication to OAuth (so ChatGPT opens our sign-in). Edit with ChatGPT
Customize → Connectors → + → Add custom connector. Paste the address, then sign in. Edit with Claude
In Kazi (app.apifunnel.ai), ask it to connect the cuttingfloor MCP server at this address, then sign in. Edit with Kazi
Then send your video: attach it in the chat in ChatGPT, or your AI gives you a link to send it from your phone or computer.
Claude Code, Codex, Cursor and any other MCP app: every setup
On your phone? Most apps add cuttingfloor on a computer. Try it now in Kazi, free
Kazi works on your phone and opens with a sample clip and edits it on cuttingfloor.
Or set it up later: Send me the setup
The agent raises the level, evens out loud and quiet parts, and measures the result so it lands at the loudness YouTube and TikTok expect without clipping.
The agent runs noise reduction and voice cleanup on the voice, and can split the voice from music or room sound, then measures the loudness so it ends at a level that sounds right.
Give it the script. The agent makes the voice with a time for every word, lays it in, and can caption it and time the cuts to it.