How to Configure MCP in the Admin Panel

The Model Context Protocol (MCP) is a powerful way to expose your APIs as tools that AI assistants can use. This guide will walk you through configuring an MCP server in the admin panel.

What is an MCP?

An MCP (Model Context Protocol) server acts as a bridge between AI assistants and your APIs. It transforms your REST API endpoints into tools that can be called by AI models through a standardized protocol.

Creating Your First MCP

Step 1: Navigate to MCPs Section

1. Log in to the admin panel
2. Navigate to the MCPs section from the sidebar
3. Click the Create MCP button

Step 2: Configure Basic Settings

When creating an MCP, you'll need to provide:

• Name: A descriptive name for your MCP server (e.g., "Weather API MCP" or "Customer Data MCP")
• Slug: A URL-friendly identifier that will be used to access your MCP

- The slug is auto-generated from the name, but you can customize it - Must be lowercase with hyphens only (e.g., weather-api-mcp) - This slug will be used in the API endpoint: /api/mcp/[slug]

• Enable/Disable: Toggle to enable or disable the MCP

- When disabled, the MCP will appear as non-existent when accessed via API

Step 3: Understanding Slugs

The slug is crucial because it's how clients will access your MCP server. For example:

• If your slug is weather-api, clients will connect to: /api/mcp/weather-api
• The slug must be unique within your account
• Once created, the slug should remain stable to avoid breaking client connections

Step 4: Save Your MCP

After filling in the details:

1. Click Create MCP to save
2. Your MCP will now appear in the MCPs list
3. You can edit or delete it at any time

Next Steps

After creating your MCP, you'll want to:

1. Create Tools: Define the tools that your MCP will expose
2. Create APIs: Set up the API endpoints that your tools will call
3. Map Tools to APIs: Connect your tools to their corresponding API endpoints

Best Practices

• Use descriptive names: Make it clear what your MCP does
• Choose meaningful slugs: Use slugs that reflect the purpose (e.g., customer-crm-api not mcp-1)
• Start with enabled: Keep your MCP enabled unless you're testing or troubleshooting
• Document your MCP: Add descriptions to help you remember what each MCP does

Troubleshooting

MCP not found error?

• Check that the MCP is enabled
• Verify the slug matches exactly (case-sensitive)
• Ensure you're using the correct API endpoint format

Can't create MCP?

• Ensure the slug is unique
• Check that the slug follows the format (lowercase, hyphens only)
• Verify you have the necessary permissions

Conclusion

Configuring an MCP is the first step in exposing your APIs to AI assistants. Once configured, you can add tools and map them to your APIs to create powerful integrations.

For more information, see our guides on creating APIs and MCP to API mapping.