Supercut Platform API and MCP docs:
You’ll need a Supercut API token before you can connect MCP.
1. Choose the right API token
Supercut supports two token types for MCP: Workspace API tokens and Personal API tokens.
Workspace API token
Use a Workspace API token if you want MCP to access recordings in the workspace when you already have the Supercut share ID.
Workspace API tokens can only be created and managed by workspace admins, scoped to the workspace.
Personal API token
Use a Personal API token if you want your AI tool to access recordings from your own Supercut account, including recordings you own, recordings specifically shared with you, or recordings you’ve watched.
Personal API tokens let you choose specific scopes, such as:
List recordings
List stacks
Read stack metadata
Read stack recordings
Read transcripts
Read video frames
Read comments
Read reactions
Read form responses
For most MCP setups, select all scopes unless you want to restrict what your AI tool can access.
2. Create your API token
Create a Workspace API token
In Supercut, open your workspace settings https://supercut.ai/workspaces
a) Go to Workspace API tokens.
b) Click Create token.
c) Give the token a name, such as: Claude MCP
d) Choose an expiration date.
e) Click Create token.
f) Copy the token and save it somewhere secure.
You’ll only see the full token once.
Create a Personal API token
In Supercut, open your workspace settings https://supercut.ai/workspaces
a) Go to Personal API tokens.
b) Click Create token.
c) Give the token a name, such as: Claude MCP
d) Choose an expiration date.
e) Select the scopes you want this token to grant.
For full MCP access, select all available scopes.
f) Click Create token.
g) Copy the token and save it somewhere secure.
You’ll only see the full token once.
3. Setting up the MCP on your agent tool
Choose the client you use and copy the relevant prompt/code:
https://api.supercut.ai/api/platform/v1/docs/#description/mcp-client-setup
Supercut uses remote HTTP MCP. Use this MCP server URL:
https://mcp.supercut.ai/mcp
Use your API token as a bearer token:
Authorization: Bearer <token>
Replace <token> or sk_your_api_key in the examples with your Supercut API token.
4. Test the connection
Open your AI tool and ask it to use Supercut.
For example if you are using a personal token:
Use the Supercut MCP to list recent recordings i have access to and extract any bug reports to work on
Or, if you have a specific Supercut share ID and using the either the workspace or personal token:
Fetch the transcript and comments for this Supercut: <share id>
If the connection is working, your AI tool should be able to retrieve Supercut data through MCP.
That's it!
Your AI tool can now read your Supercuts directly. Ask it to pull transcripts, summarise recordings, surface comments, or extract action items from any video you have access to.