Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@capcut-mcp Add auto captions to my latest draft and send me a review link"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Yes, ChatGPT, Claude and other AI agents can work in your CapCut account. Cotlet lets your agent export drafts, add auto captions, send review links and manage team spaces in your own CapCut account.
Export my Launch teaser draft as a 1080p MP4.
Add auto captions to the latest draft and send me a review link.
List the editors in my Default space.
Search the template gallery for a summer product launch.
Invite Maya Brooks to our team space as an editor.
Related MCP server: CapCut MCP Server
Get your API key at https://cotlet.dev/app and export it as COTLET_API_KEY.
claude mcp add --transport http cotlet https://cotlet.dev/mcp/v1 --header "Authorization: Bearer $COTLET_API_KEY"Settings → Apps → add connector URL https://cotlet.dev/mcp/v1, then sign in with Cotlet.dev. No API key needed.
Copy .cursor/mcp.json from this repo into your project:
{
"mcpServers": {
"cotlet": {
"url": "https://cotlet.dev/mcp/v1",
"headers": {
"Authorization": "Bearer ${COTLET_API_KEY}"
}
}
}
}This repo is a real MCP server. It speaks MCP over stdio, lists its CapCut tools without any key, and forwards every tool call to the hosted Cotlet.dev server. The hosted URL above stays the recommended way. Use this when your client only starts local servers.
{"mcpServers":{"capcut":{"command":"uvx","args":["capcut-mcp"],"env":{"COTLET_API_KEY":"your-key"}}}}Or with Docker:
docker build -t capcut-mcp .
docker run -i --rm -e COTLET_API_KEY capcut-mcpTools: capcut_guide, capcut_status, capcut_open, capcut_read, capcut_click, capcut_type, capcut_close. Without COTLET_API_KEY a tool call answers: "Set COTLET_API_KEY. Get a key at https://cotlet.dev/app".
Python 3.10+, nothing else to install. From PyPI:
pip install capcut-mcp # or run it once: uvx capcut-mcp connect
capcut-mcp connectOr straight from this repo:
export COTLET_API_KEY="your-key"
python3 capcut_agent.py connect # ready config for Claude Code, Cursor, ChatGPT
python3 capcut_agent.py status # is CapCut ready in your Cotlet.dev account
python3 capcut_agent.py guide # CapCut screens and sign-in link
python3 capcut_agent.py read # open CapCut, read the page, close the browser
python3 capcut_agent.py tools # list Cotlet.dev tools
python3 capcut_agent.py serve # start the stdio MCP server (same as no argument)Does CapCut have an MCP server or an API?
CapCut has no public API for drafts, exports or sharing and publishes no MCP address for Claude or Cursor. Cotlet runs hosted CapCut tools over MCP and REST.
How do I connect CapCut to ChatGPT?
Sign in to CapCut in your Cotlet Cloud Browser, enable CapCut for agents, then add https://cotlet.dev/mcp/v1/connect in ChatGPT.
Can I use CapCut with Claude?
Yes. Add https://cotlet.dev/mcp/v1 as a custom connector in Claude after enabling CapCut for agents.
What can an AI agent do in CapCut?
Your agent can export drafts, add auto captions, share drafts for review, manage team spaces and search the template gallery.
Start free: https://cotlet.dev/signup?app=capcut&utm_source=github&utm_medium=readme&utm_campaign=capcut
Click a numbered element on the current CapCut page.
Inputs: element (required).
Close the CapCut browser session when the work is done, so credits stop running.
Inputs: no named parameters.
How to work in CapCut (export and share drafts): its screens and the sign-in link. Call this first.
Inputs: task.
Open a CapCut page in your signed-in Cloud Browser (default https://www.capcut.com) to work with export and share drafts.
Inputs: url.
Read the current CapCut page: its text and numbered buttons, links and fields.
Inputs: element.
Check that your Cotlet.dev account and CapCut are ready: account, credits and CapCut connection.
Inputs: no named parameters.
Type text into a numbered field on the current CapCut page.
Inputs: text (required), element (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 7 tools
The browser-action tools (open, read, click, type, close) each target a distinct step in the interaction loop, and guide vs. status are mostly separable (instructions vs. account/connection check). The only mild overlap is that both guide and status are framed as 'call first' prerequisites, which could cause a hesitation on entry.
Every tool uses the same capcut_ prefix followed by a single lowercase verb (guide, status, open, read, click, type, close). The pattern is predictable and readable throughout.
Seven tools is well-scoped for a browser-automation flow: two onboarding helpers plus the core open/read/click/type/close loop. Each tool earns its place with no redundant members.
The core loop covers opening a page, reading elements, clicking, typing and closing, but there is no scroll, wait-for-load, key press, or explicit navigate/screenshot operation. These are notable gaps for a general cloud-browser surface, though agents can partly work around them via repeated read/click.
CapCut MCP server for ChatGPT, Claude and any AI agent: list drafts, export draft, add auto captions, share draft for review, invite members to a space, set member role, transfer editing permission and search templates in your own CapCut account. Cotlet works in the account through your Cloud Browser, a real Chrome where you sign in to CapCut yourself; Cotlet does not store your password and you need no CapCut API key. About the official CapCut API: Partly. CapCut has an official plugin for the ChatGPT desktop app that builds drafts and adds captions, effects and templates (capcut.com, checked 6 Nov 2026). Setup and FAQ: https://cotlet.dev/integrations/capcut-api-mcp-server
The agentic video editor for ChatGPT and Claude. Ask in your chat and it cuts, captions, adds music, titles and motion graphics on a real timeline you can open and adjust.
AI video editor for agents and humans: timeline, captions, color, audio and generation as MCP tools.
Clip videos into captioned shorts, add captions, and schedule posts from AI agents.
