Connect custom AI tools
Your AI tool supports MCP but we don't have a guide for it yet? Use these connection details to give it full access to read and update your Kick data
What You'll Need
An MCP-compatible AI tool or IDE
A Kick account with access to at least one workspace
How to confirm your commonly used tool is MCP-compatible
Not sure if your tool supports MCP? Here's how to check:
Search their docs for "MCP" or "Model Context Protocol"
Check settings for "Connectors," "Integrations," or "Remote servers"
Ask their support — MCP adoption is growing fast; they may have it in beta
Look for recent announcements — tools are adding MCP support regularly
If your tool doesn't support MCP yet, let them know you want it. The protocol is open-source and growing.
Standard Connection Details
Use these values when configuring Kick MCP in any compatible client:
Name
Kick
MCP server URL
https://use.kick.co/mcp
Transport
Streamable HTTP, HTTP, or Remote MCP, depending on the client's wording
Authentication
OAuth when available, otherwise API key or bearer token
Bearer header
Authorization: Bearer kick_pat_...
Connection Steps
The exact steps vary by client, but the general pattern is:
Step 1: Find MCP or Connector Settings
Look for settings related to:
MCP servers
Custom connectors
Remote servers
Tools or integrations
Common locations:
Settings → Connectors
Settings → Tools & MCP
Settings → Integrations
Preferences → Extensions
Step 2: Add the Kick Server
Click Add server, Add connector, or similar
Enter the connection details from the table above
Choose OAuth for authentication if available
Save or add the configuration
Step 3: Authorize (OAuth) or Add Token
If using OAuth:
A browser window will open to Kick's authorization page
Sign in to your Kick account if prompted
Choose your permissions and workspace access
Click Authorize
If using bearer token:
Generate a personal access token in Kick (Settings → API)
Copy the token (it starts with
kick_pat_)Paste it in the authentication field in your MCP client
Step 4: Restart or Reconnect
After adding the server:
Restart the client application, or
Reconnect to reload the MCP server list
Verify Your Connection
Test that everything is working by asking your MCP client:
The client should respond with:
A list of available Kick tools (like
transactions_find,reports_profit_loss_get,workspaces_list)Confirmation of which tools are read-only
The name of your active workspace
Common Authentication Methods
OAuth (Recommended)
OAuth automatically handles authorization through your browser. Most modern MCP clients support this method.
Advantages:
No manual token management
Automatic token refresh
Easier revocation
When to use: Choose OAuth if your client supports it.
Bearer Token / API Key
Bearer tokens are personal access tokens generated in Kick and manually added to your MCP client.
Advantages:
Works in clients without OAuth support
More granular access control
Useful for automation
When to use: Choose bearer token if OAuth isn't available or you need token-based authentication.
To generate a bearer token:
Log in to Kick
Go to Settings → API
Click Generate MCP token
Copy the token (starts with
kick_pat_)Store it securely — it won't be shown again
Troubleshooting
Connection Failed or No Tools Showing
The server configuration didn't load successfully.
Verify the MCP server URL is exactly:
https://use.kick.co/mcpCheck that authentication is configured correctly
Restart or reconnect the client
Check client documentation for MCP-specific requirements
Authorization Failed (OAuth)
The authorization didn't complete successfully.
Make sure you're logged in to your Kick account
Complete the full authorization flow
Click Authorize on the Kick authorization page
Try disconnecting and reconnecting the server
Invalid Token (Bearer Token)
The personal access token isn't working.
Verify the token was copied correctly (no extra spaces)
Check that the token hasn't expired in Kick settings
Generate a new token if needed
Make sure the bearer header format is:
Authorization: Bearer kick_pat_...
Wrong Workspace Showing
Your Kick account has access to multiple workspaces. Tell the client which workspace you want to use:
Data Privacy and Security
Your data stays in Kick: MCP clients request data through the connection but don't store your Kick data permanently
Workspace permissions apply: Clients can only access workspaces you already have permission to view in Kick
You control access: You can disconnect Kick or revoke tokens at any time
Confirmation required for changes: Write operations require your explicit approval before execution
Next Steps
Now that you're connected:
Try a sample query: "List my workspaces" or "What tools can you use?"
Explore reports by asking your AI tool to generate financial reports for your clients
Check out the Kick MCP Use Case Library for advanced use cases
Last updated
Was this helpful?
