TypeScript

Prerequisites

Node.js ^20.19.0 or >=22.12.0
An MCPFY API key - see Get your API key if you don’t have one yet

There are three ways to use mcpfy-pulse. None of them edit your files for you - pick the one that matches your situation.

Using mcpfy-sdk
Your own server (raw SDK)
Someone else's server (no source)

No install step - mcpfy-pulse ships bundled with mcpfy-sdk. Just set one environment variable and restart your server however you normally do:

MCPFY_API_KEY=mk_live_xxx node dist/server.js

mcpfy-sdk checks for MCPFY_API_KEY internally and wraps its own transport automatically. Unset the variable and nothing changes - no code path is even touched.

For anyone who wrote their own server on the raw @modelcontextprotocol/sdk (or anything else that exposes a Transport) and has the source in front of them.

Install the package

npm install mcpfy-pulse

Set your API key

In your .env file:

MCPFY_API_KEY=mk_live_xxx

Wrap your transport

Right before you connect it. This works for any transport - stdio, HTTP, SSE - not just the stdio example below:

import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";
import { withMcpfyTelemetry } from "mcpfy-pulse";

const transport = new StdioServerTransport();
await server.connect(
  withMcpfyTelemetry(transport, { apiKey: process.env.MCPFY_API_KEY })
);

withMcpfyTelemetry wraps the Transport’s onmessage/send seam - the two points every JSON-RPC message passes through regardless of which SDK built the server. If apiKey is unset, it returns the original transport unchanged, so it’s safe to leave this in place across environments.

No install step - npx fetches mcpfy-proxy automatically the first time it runs. Edit your MCP client’s config (claude_desktop_config.json, Cursor’s mcp.json, etc.) to route the command through the proxy:

// before:
"github": { "command": "npx", "args": ["-y", "@modelcontextprotocol/server-github"] }

// after:
"github": {
  "command": "npx",
  "args": ["-y", "mcpfy-proxy", "--", "npx", "-y", "@modelcontextprotocol/server-github"],
  "env": { "MCPFY_API_KEY": "mk_live_xxx" }
}

mcpfy-proxy becomes the process your client spawns. It spawns the real command as its own child, sits in that child’s stdin/stdout, and forwards every byte unchanged while classifying JSON-RPC messages on the side. Works for any language - Python, Go, Rust, anything - since it only ever reads newline-delimited JSON off a pipe.

Advanced options

withMcpfyTelemetry and the mcpfy-sdk integration both accept the same options object, and every field falls back to an environment variable if omitted:

Option	Env var	Default
`apiKey`	`MCPFY_API_KEY`	(required to send data)
`endpoint`	`MCPFY_TELEMETRY_ENDPOINT`	`https://api.mcpfy.ai/v1/telemetry/ingest`
`flushIntervalMs`	-	`5000`
`maxBatchSize`	-	`500`

What’s next

Once your server sends its first event, head to the dashboard to see it.

View telemetry in the dashboard

See tool-level usage, health scores, and quality signals once data starts flowing.

Getting started

MCPfy Cloud

MCPfy SDK

Pulse SDK

Prerequisites

Advanced options

What’s next

View telemetry in the dashboard

​Prerequisites

​Advanced options

​What’s next

View telemetry in the dashboard

Prerequisites

Advanced options

What’s next