CapCut MCP Server
Drafts, exports, captions, sharing and team spaces in CapCut.
Hosted MCP Server
npx add-mcp 'https://cotlet.dev/mcp/v1/capcut'Installs into Claude Code, Codex, Cursor and more
Documentation
Cotlet · Integrations · How it works · Security · For developers · For automation agencies · Blog · API docs · Help center · Changelog · Contact · Pricing · Sign up
CapCut MCP server: Let your AI agent export and share your CapCut drafts.
CapCut has no public API for drafts, exports or sharing and publishes no MCP address for Claude or Cursor (checked 6 Nov 2026). Its official AI route is the CapCut plugin for the ChatGPT desktop app on PC, which builds and edits drafts inside a ChatGPT task. Cotlet runs 13 CapCut tools in your own account, hosted, over MCP and REST.
Connect CapCut to Claude, ChatGPT, n8n or a script. Your agent exports drafts, adds auto captions, sends review links and manages team spaces in your own CapCut account. Cotlet hosts it, so there is no developer setup and no server to run.
- Free 7-day trial, no card
- Not affiliated with CapCut
- Page: https://cotlet.dev/integrations/capcut-api-mcp-server
What can your AI agent do in CapCut?
- Export drafts at a set size and format
- Add auto captions to a draft
- Share a draft for review
- Invite an editor to a team space
- Check who can edit your spaces
- Search the template gallery
Does CapCut have an MCP server or an API?
- The plugin generates media, builds an editable timeline and adds captions, effects and templates in chat. CapCut says it is made for the ChatGPT desktop app on PC, available outside the United States, with the U.S. coming soon.
- CapCut links no developer portal, API key or webhooks documentation. Zapier and n8n have no CapCut app page.
- Community servers on GitHub mostly write draft files on your own computer. You open them in CapCut and run the export there.
| What you need | CapCut plugin for ChatGPT | Community MCP server | Cotlet |
|---|---|---|---|
| Export at a chosen size and format | Finish in the editor | Draft file, export in CapCut | Yes |
| Review links, spaces and members | Not listed by CapCut | No | Yes, links after your OK |
| Works in Claude, n8n or a script | ChatGPT desktop only | Local MCP clients | Yes |
| Runs on a schedule, 24/7 | Inside a ChatGPT task | While your computer is on | Hosted |
| No install, no key, no code | Install from a prompt | Clone and run it | Sign in once |
Checked 6 Nov 2026. Sources: CapCut: CapCut × ChatGPT plugin, CapCut newsroom: CapCut World Los Angeles, GitHub: VectCutAPI, CapCut: Collaborate on CapCut online
How does Cotlet connect to CapCut?
- Sign in to CapCut. Use your own account in its Cotlet Cloud Browser.
- Choose MCP or REST. Connect your agent or create a Cotlet API key.
- Run your workflow. Ask your agent or send an HTTP request.
We never keep your CapCut password, and by default no review link or invite goes out until you approve it.
Connect CapCut to your agent or workflow
Use MCP with your AI agent, or HTTP requests in n8n, Make, Zapier and scripts. Follow the MCP setup below or jump to the REST guide.
- Enable CapCut for agents. Sign up, log in to CapCut in its Cloud Browser, then press Enable for agents.
- Add the Cotlet address. Add https://cotlet.dev/mcp/v1 as an MCP server or connector in your agent.
- Approve the connection. Chat apps ask you to sign in to Cotlet. Editors, n8n and scripts use an API key from the console.
One shared address, https://cotlet.dev/mcp/v1, also carries CapCut next to your other apps. If you enable CapCut after connecting, refresh the connector so the new tools show up.
Claude:
Settings > Connectors > Add custom connector
URL: https://cotlet.dev/mcp/v1/connect
Claude Code:
claude mcp add --transport http cotlet-capcut \
https://cotlet.dev/mcp/v1 \
--header "Authorization: Bearer $COTLET_API_KEY"
ChatGPT:
Settings > Apps & Connectors > Create
MCP server URL: https://cotlet.dev/mcp/v1/connect
Authentication: OAuth
Cursor:
// ~/.cursor/mcp.json
{
"mcpServers": {
"cotlet-capcut": {
"url": "https://cotlet.dev/mcp/v1",
"headers": { "Authorization": "Bearer ${env:COTLET_API_KEY}" }
}
}
}
Codex:
# ~/.codex/config.toml
[mcp_servers.cotlet-capcut]
url = "https://cotlet.dev/mcp/v1"
bearer_token_env_var = "COTLET_API_KEY"
n8n:
MCP Client Tool (inside an AI Agent)
Transport: HTTP Streamable
URL: https://cotlet.dev/mcp/v1
Auth: Bearer, <your key>
Tool: capcut_read
Make, Zapier:
Make: MCP Client > Call a tool
Zapier: MCP Client by Zapier > Run Tool
Server URL: https://cotlet.dev/mcp/v1
Token: <your key>
Tool: capcut_write
How to use the CapCut REST API
Create your Cotlet API key
Sign in to your CapCut account in its Cotlet Cloud Browser and enable the app for agents. Managing members and roles needs owner or admin rights in the space.
Create a Cotlet API key with access to CapCut and its read operations. Store it as COTLET_API_KEY in your environment or automation platform's credential store.
Make your first CapCut API request
cURL: list_drafts:
curl https://cotlet.dev/api/v1/capcut/drafts \
-H "Authorization: Bearer $COTLET_API_KEY"
Read drafts and continue the list
A successful REST call puts the tool output in result. This illustrative response excerpt omits call IDs, messages and credit details.
Illustrative response excerpt:
{
"status": "success",
"app": "capcut",
"tool": "list_drafts",
"result": {
"drafts": [
{
"draft_id": "dr_5c7e10",
"name": "Launch teaser",
"editor": "Maya Brooks",
"updated_at": "2026-11-04T16:20:00Z"
}
],
"next_cursor": null
}
}
Read result.drafts and pass result.next_cursor unchanged as cursor until it is null. Without space_id the call reads your Default space; pass a space_id from list_spaces for the others.
Keep draft IDs in your workflow so later runs match the same drafts. Only one member edits a draft at a time, so editor tells you who holds it.
CapCut workflows for agents and automations
| Tool | Connect through | Example workflow |
|---|---|---|
| n8n | HTTP Request | Read list_drafts for a space every Monday and post the draft names and editors to your team channel. |
| Python or Node.js | REST API | Call list_space_members for each space from list_spaces and write a roles report to a sheet. |
| Make | HTTP > Make a request | Call list_share_history with direction set to shared_by_you and keep a list of the review links that are out. |
Send Authorization: Bearer <your Cotlet API key> in request headers. You configure the schedule, destination and record comparisons.
Cotlet reads CapCut; your workflow steps write to your chosen destination. These HTTP workflows do not require a native Cotlet connector or a prebuilt template.
The API and MCP reference lists operations and inputs. The REST API guide explains authentication, errors and approvals.
CapCut API endpoints and MCP tools
Team spaces and members
CapCut allows 5 spaces per user, Default space included, and 30 members per space (CapCut help, Aug 2026).
- List spaces (
list_spaces, read): Your CapCut spaces with your role and member count. - List space members (
list_space_members, read): Members of a space with their role. - Get space invite link (
get_invite_link, read): The invitation link of a space, the one behind Copy in Invite members. - Create space (
create_space, write, asks first): Creates a new team space in your Workspace. - Invite members to a space (
invite_members, write, asks first): Sends space invitations by email through Invite members. - Set member role (
set_member_role, write, asks first): Sets a member's access in a space to admin or collaborator.
Drafts, captions and export
Export goes to the Cloud Browser's downloads and reaches no one. 2K and 4K depend on your device and app version.
- List drafts (
list_drafts, read): Drafts in a space with who holds editing permission and the last edit. - Add auto captions (
generate_captions, write): Runs auto captions on a draft and adds the lines to its timeline. - Export draft (
export_draft, write): Exports a draft with the file name, resolution, quality, frame rate and format you set. - Search templates (
search_templates, read): Templates from the CapCut gallery that match a search.
Review links and editing permission
Only one member edits a draft at a time, so a transfer turns the previous editor to view only.
- List share history (
list_share_history, read): Drafts shared with you or by you, with the review link. - Share draft for review (
share_for_review, write, asks first): Shares a draft through Export > Share for review so others can comment. - Transfer editing permission (
transfer_editing_permission, write, asks first): Hands editing permission on a draft to another team member.
Inputs, REST path and example response of every tool: https://cotlet.dev/integrations/capcut-api-mcp-server#tools
Who can see my CapCut session, and how do I revoke it?
- What we store. Your logged-in CapCut profile, encrypted with AES-256-GCM, in Finland. Not the password.
- What your agent gets. The 13 CapCut tools. Never the cookie.
- What the session sees. Your drafts, spaces, members and share history.
- What no tool touches. Billing, plan, password, two-factor and deleting drafts.
- If CapCut signs you out. Your agent gets a link to sign in again. You do it, we never do.
- How to cut access. Disable for agents or reset the browser in the Cotlet console.
Our site scores A+ on Qualys SSL Labs, securityheaders.com and Mozilla Observatory (live results). There is no SOC 2 report or third-party pentest yet.
What happens when CapCut changes?
- 12 h between tests of the CapCut tools
- $0 for a call to a CapCut tool paused for repair
When CapCut moves a button or renames a menu, the fix happens on our side and your prompts keep working. status.cotlet.dev watches the site, console and API every 3 minutes.
Approvals, undo and mistakes
- Finds the draft. Looks it up by name in the space you name.
- Sets the visibility. Everyone with the link, or the space only.
- Waits for your OK. The share shows up in your Cotlet console first.
- Reports the link. Review link, visibility and who got an email.
| Action | Undo |
|---|---|
| Export a draft | Delete the file; the draft is unchanged |
| Add auto captions | Delete the caption lines in the editor |
| Share a draft for review | No revoke tool here; list_share_history shows which links are out |
| Invite members | The email is out; no tool removes a member |
| Set a member role | Set the old role again |
| Transfer editing permission | Take it back the same way |
| Create a space | No, CapCut does not support deleting a space |
5 of 13 tools ask first: the ones that email or invite someone, change access or create a space. Exports and captions reach no one, so they run at once. Switch the wait off per tool for a scheduled run.
Create a space is permanent. You get 5 spaces, Default included.
MCP call:
{
"name": "share_for_review",
"arguments": {
"draft_id": "dr_5c7e10",
"visibility": "space",
"emails": [
"anna@example.com"
]
}
}
REST:
curl -X POST https://cotlet.dev/api/v1/capcut/drafts/dr_5c7e10/review/share \
-H "Authorization: Bearer $COTLET_API_KEY" \
-H "Content-Type: application/json" \
-d '{"visibility":"space","emails":["anna@example.com"]}'
Tool output:
{
"draft_id": "dr_5c7e10",
"visibility": "space",
"shared_with": [
"anna@example.com"
],
"url": "https://www.capcut.com/share/example-review-link"
}
Scripts and cron jobs reach the same tools over REST at /api/v1/capcut, with the same key and price. A setting CapCut refuses, such as 4K on a device without it, fails with CapCut's message and buys nothing. A failed call costs nothing. The tools table lists the REST path of each tool.
How much does the CapCut MCP server cost?
About $12 a month for a small video team exporting 12 drafts a month, 2,120 credits. Your CapCut plan is billed by CapCut as before.
| Job | Calls | Credits |
|---|---|---|
| Auto captions, then export a draft | 3 × 8 | 240 |
| Export a draft as it is | 2 × 4 | 80 |
| Share a draft for review | 2 × 6 | 120 |
| Weekly look at spaces and members | 3 × 4 | 120 |
| Invite an editor and set the role | 3 × 2 | 60 |
| Cloud Browser, 30 days | 1,500 | |
| Total | 2,120 |
- Free 7-day trial, no card
- 10 credits per successful call, MCP or REST
- 50 credits a day for the Cloud Browser
- CapCut lists Free, Pro, Ultra and Teams plans, billed by CapCut
The month is our example: one client-review space of about 4 members, 12 drafts exported, 6 review shares, 2 new editors. No public figure for a typical CapCut team exists. Dollars at the $29 pack rate.
CapCut MCP server on GitHub, the ChatGPT plugin or hosted: which to use
| Option | What it does | Where it stops |
|---|---|---|
| CapCut × ChatGPT plugin | Official. Builds a draft from a prompt or upload, edits the timeline, captions and effects in chat. | ChatGPT desktop app on PC. CapCut lists export, sharing and space management in the editor, not in the plugin. |
| sun-guannan/VectCutAPI | HTTP API and MCP mode that write CapCut and Jianying draft files. 2,362 stars, last commit 24 Oct 2026. | Runs on your computer with Python and FFmpeg. Its feature list has no spaces, sharing or members. |
| Hommy-master/capcut-mate | FastAPI draft service that works with Coze and n8n. 2,108 stars, last commit 1 Nov 2026. | Chinese first. You host it, and it covers drafts, not your team spaces. |
| mrbuslov/capcut-ai-editor | MCP server that cuts silences and repeated takes using CapCut's auto captions. 123 stars, last commit 27 Oct 2026. | One editing job on a local project. |
| Atx-Guy/capcut-mcp-server | MCP wrapper for the VectCutAPI backend. 90 stars, last commit 13 Jan 2026. | Needs that backend running on your computer. |
Use CapCut's plugin when you shape the cut in ChatGPT. Use a draft writer when you build timelines from code on your own machine. Use Cotlet when the chores around the edit are the job: exports, captions, review links and who can edit, from Claude, ChatGPT or a schedule, with nothing to host.
Who runs Cotlet
Take Group sp. z o.o. runs Cotlet. It is a software company registered in Poland, EU (VAT PL6342984273), working since 2019, founded by Bart Krawczyk (Co-founder, CTO) and Jake Slugocki (Co-founder, SEO, sales and marketing). Write to contact@cotlet.dev with questions about the CapCut tools.
CapCut MCP and API: questions
How do I connect CapCut to ChatGPT or Claude?
Sign up at cotlet.dev, log in to CapCut once in its Cloud Browser and add Cotlet to ChatGPT or Claude, following the steps in your console. No API key is needed. ChatGPT and Claude then work in your CapCut account through Cotlet, in their phone apps too.
What can ChatGPT do in CapCut?
With Cotlet, ChatGPT, Claude and other AI agents handle jobs like these in CapCut: export drafts at a set size and format; add auto captions to a new draft; send a draft for client review. You ask in plain words, and anything that reaches other people waits for your OK first.
Can I use the CapCut API without an AI agent?
Yes. Call Cotlet's REST API from a script or an HTTP step in n8n, Make or Zapier with your Cotlet key. MCP is the alternative for AI agents; you do not need to connect an agent to use REST.
How do I get access?
Create a free account at https://cotlet.dev, open CapCut under Integrations in the console and log in to CapCut in its Cloud Browser. The trial starts when you sign up, no card asked.
Can my agent edit the timeline, cut clips or add effects?
No. The tools cover what sits around the edit: drafts, auto captions, export, review links, team spaces and template search. Cutting, effects and transitions stay in the editor, where you or your team do them by hand.
Do I need CapCut Pro for exports?
Not for the call itself. CapCut's help says 2K and 4K depend on your device, the app version and the platform, and the project resolution must be set before editing. When the app refuses a setting or asks for a plan, export_draft reports it and changes nothing else.
Does CapCut have webhooks?
CapCut documents none (checked 6 Nov 2026). Nothing tells an agent that a draft changed. Run your agent on a schedule instead, for example a weekly list_drafts and list_share_history pass.
Is the MCP server different from the REST API?
Only the door is different. Both expose the same 13 tools, inputs and answers, with one API key and one price. Agents use MCP; scripts and HTTP steps use REST.
Is Cotlet affiliated with CapCut?
No. Cotlet is independent and not affiliated with CapCut or ByteDance.
Give your AI agents your signed-in browser.
Product: Integrations · How it works · Pricing · Security
Resources: Blog · API docs · Help center · Changelog · GitHub
Solutions: For developers · For automation agencies
Company: About · AI info: company facts · Contact
Legal: Terms of service · Privacy policy · Refund policy · Acceptable use · Cookie policy · DPA · Subprocessors
Elsewhere: Status page · Cotlet on GitHub · Cotlet on X · Cotlet on LinkedIn · Cotlet on Product Hunt · Cotlet on Crunchbase · Cotlet on SaaSHub
Cotlet.dev is built by Bart Krawczyk and Jake Slugocki, the founders of Take Group sp. z o.o., ul. Wincentego Janasa 4, 40-855 Katowice, Poland. KRS 0000871398, NIP 6342984273, VAT EU: PL6342984273. About us
Cotlet is an independent service. It is not affiliated with, endorsed by or sponsored by any of the products listed on this site. Product names are trademarks of their respective owners.