​
Quick Start Guide

Get up and running with Shinzo Platform in just a few minutes. This guide will walk you through instrumenting your MCP server and viewing telemetry data in the dashboard.

​
Prerequisites

  • Node.js 18 or higher
  • An existing MCP server using @modelcontextprotocol/sdk
  • A Shinzo Platform account (sign up at app.shinzo.ai)

​
Step 1: Install the SDK

Install the Shinzo instrumentation SDK in your MCP server project: 
npm install @shinzolabs/instrumentation-mcp

​
Step 2: Get Your Ingest Token

  1. Sign up or log in at app.shinzo.ai
  2. Navigate to Settings → Ingest Tokens
  3. Copy your ingest token (it will look like abc123def456...)

​
Step 3: Instrument Your Server

Add telemetry to your MCP server with just a few lines of code: 
import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"
import { instrumentServer } from "@shinzolabs/instrumentation-mcp"

const server = new McpServer({
  name: "my-awesome-server",
  version: "1.0.0",
  description: "My MCP server with telemetry"
})

// Add this after creating your server
const telemetry = instrumentServer(server, {
  serverName: server.name,
  serverVersion: server.version,
  exporterEndpoint: "https://api.app.shinzo.ai/telemetry/ingest_http",
  exporterAuth: {
    type: "bearer",
    token: "your-ingest-token-here" // Replace with your actual token
  }
})

// Continue with your existing server setup
server.tool("example_tool", {
  description: "An example tool",
  inputSchema: {
    type: "object",
    properties: {
      message: { type: "string" }
    }
  }
}, async (args) => {
  return { content: `Hello, ${args.message}!` }
})
Replace your-ingest-token-here with your actual ingest token from Step 2.

​
Step 4: Start Your Server

Run your instrumented MCP server as usual. Telemetry data will automatically be collected and sent to Shinzo Platform whenever:
  • Tools are executed
  • Resources are accessed
  • Errors occur
  • The server starts or stops

​
Step 5: View Your Data

  1. Go to app.shinzo.ai
  2. Navigate to the Dashboard to see real-time metrics
  3. Check the Traces page to see individual tool executions
  4. Visit Resources to monitor server performance

​
Example: Complete MCP Server

Here’s a complete example of an instrumented MCP server: 
import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"
import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"
import { instrumentServer } from "@shinzolabs/instrumentation-mcp"

const server = new McpServer({
  name: "weather-server",
  version: "1.0.0",
  description: "Weather information MCP server"
})

// Instrument the server
const telemetry = instrumentServer(server, {
  serverName: "weather-server",
  serverVersion: "1.0.0",
  exporterEndpoint: "https://api.app.shinzo.ai/telemetry/ingest_http",
  exporterAuth: {
    type: "bearer",
    token: process.env.SHINZO_TOKEN
  }
})

// Add a weather tool
server.tool("get_weather", {
  description: "Get weather for a city",
  inputSchema: {
    type: "object",
    properties: {
      city: { type: "string" }
    },
    required: ["city"]
  }
}, async (args) => {
  // Simulate API call
  const weather = {
    city: args.city,
    temperature: Math.round(Math.random() * 30 + 10),
    condition: "sunny"
  }

  return {
    content: `Weather in ${weather.city}: ${weather.temperature}°C, ${weather.condition}`
  }
})

// Start the server
async function main() {
  const transport = new StdioServerTransport()
  await server.connect(transport)
}

main().catch(console.error)

​
What Gets Collected?

By default, Shinzo Platform collects: ✅ Tool execution metrics: Response times, success/failure rates ✅ Resource access patterns: Which resources are accessed and how often ✅ Server health metrics: Memory usage, request rates ✅ Error tracking: Exceptions and error rates ❌ Tool arguments: Not collected by default (privacy-first) ❌ Response data: Not collected by default ❌ PII: Automatically sanitized when detected

​
Next Steps

Now that your server is instrumented:

Configure SDK

Learn about advanced configuration options including PII sanitization and custom processors

Explore Dashboard

Discover all the insights available in the Shinzo Platform dashboard

​
Troubleshooting

​
Not seeing data?

  1. Check your ingest token: Make sure it’s correct and active
  2. Verify network access: Ensure your server can reach api.app.shinzo.ai
  3. Check console logs: The SDK logs important information about telemetry export
  4. Test with console exporter: Use exporterType: "console" to see telemetry locally

​
Performance concerns?

The SDK is designed to be lightweight with minimal performance impact:
  • Telemetry is exported in batches
  • Network calls are non-blocking
  • Sampling can be configured to reduce data volume
  • Memory usage is optimized for production use
Need help? Contact us at austin@shinzolabs.com