Edit a video by asking AI in PandaStudio
PandaStudio has a built-in AI agent you direct in plain language; it performs real timeline edits (cuts, captions, zooms, motion graphics) and runs on your own ChatGPT, Claude, Gemini, or Kimi account, so there is no AI markup.
You describe the outcome and the agent does the editing. It understands the project, the transcript, and the full editor, so it can trim rambling, add word-synced captions, place zooms on the right beats, build motion graphics, and edit in the measured style of a top creator. Because it uses your own AI account, you pay your provider directly with no cap or margin from us, and the same capabilities are reachable from Claude Code, Cursor, Cline, Continue, and Claude Desktop through a CLI and MCP server.
In the app
- 1Open the AI tab (or the floating chat bubble) in the editor.
- 2Connect your ChatGPT, Claude, Gemini, or Kimi account once.
- 3Type what you want in plain language and watch the edits land on the timeline.
- 4Iterate: ask for changes, undo, or take over by hand at any point.
Ask the AI
Open the AI chat in the editor and say any of these. The agent does the edits for you.
CLI & MCP
Every capability is a verb, reachable from the terminal or from an MCP agent (Claude Code, Cursor, Cline, Continue, Claude Desktop).
pandastudio project.apply-edit-plan --id=$ID --plan='[...]'
150+ verbs cover the whole editor; agents call them directly.
npx @writepanda/mcp
Exposes every verb to Claude Code, Cursor, Cline, Continue, Claude Desktop.