Getting Started
What it takes to turn your API into an AI-ready MCP service.
Prerequisites
- API credentials for the service you want to MCPify
- Basic understanding of your API's endpoints
- Access to MCPify gateway (request access if needed)
Step 1: Create Your Configuration
Create a JSON configuration file that describes your API. This tells MCPify how to interact with your service.
{
"service_name": "your-api",
"base_url": "https://api.your-service.com/v1",
"auth_type": "bearer",
"tools": {
"test_connection": {
"description": "GET /status - Test API connection\n\nVerifies credentials and connectivity",
"endpoint": "/status",
"method": "GET"
},
"list_items": {
"description": "GET /items - List all items with pagination",
"endpoint": "/items",
"method": "GET",
"input_schema": {
"type": "object",
"properties": {
"limit": {"type": "integer", "default": 10},
"offset": {"type": "integer", "default": 0}
}
}
}
}
}Step 2: Send it to us
MCPify is a hosted gateway — we register the service against your configuration rather than you deploying anything.
Start here: tell us about the API and send the configuration with it.
Credentials: once the integration is agreed, submit them at credentials.mcpify.org — they go to Google Secret Manager, never into the config file.
Step 3: Test Your MCPified API
Your API is now available as an MCP service! Test it using the MCP protocol:
# Test your MCPified API
curl -X POST https://your-api.mcp.mcpify.org/mcp \
-H "Content-Type: application/json" \
-H "X-API-Key: your-gateway-api-key" \
-d '{
"jsonrpc": "2.0",
"method": "tools/list",
"id": 1
}'Expected response:
{
"jsonrpc": "2.0",
"result": {
"tools": [
{
"name": "test_connection",
"description": "Test API connection and verify credentials",
"inputSchema": {"type": "object", "properties": {}}
},
{
"name": "list_items",
"description": "List all items with pagination",
"inputSchema": {
"type": "object",
"properties": {
"limit": {"type": "integer", "default": 10},
"offset": {"type": "integer", "default": 0}
}
}
}
]
},
"id": 1
}Step 4: Connect to AI Assistants
Your MCPified API can now be used with any MCP-compatible AI assistant:
Claude Desktop
Add your service URL to Claude's MCP configuration file
Custom Integration
Use the MCP client library to integrate with your own applications
What's Included
When you MCPify an API, you automatically get:
✅ Token Counting
Automatic token counting, with responses cached per tool
✅ Smart Caching
Response caching with field filtering support
✅ OAuth Management
Secure token storage with automatic refresh
✅ Rate Limiting
Built-in protection against API abuse
✅ Analytics
Usage tracking and performance metrics
✅ Data Navigation
Tools for JSON manipulation and exploration
Next Steps
- Read about our Philosophy to understand our approach
- Explore Data Navigation tools
- Learn about Writing effective tool descriptions
- Check the API Reference for advanced features