mcpField
S
server-track-lab

server-track-lab

Local
github.com/Surface-opsBusiness & Marketing

About server-track-lab

server-track-lab lets AI agents send server-side conversion events to ad platforms using a server-track-lab API key. Meta events are forwarded live, while Google is validate-only and LinkedIn runs as a dry run.

Key features

  1. 01Send conversion events to Meta
  2. 02Normalize and hash event data
  3. 03Deduplicate events by ID
  4. 04Test Meta event connections
  5. 05Review monthly event statistics

Use cases

  1. 01Send lead or purchase conversions
  2. 02Check a Meta pixel connection
  3. 03Inspect accepted and failed events

MCP support

Provides MCP tools for tracking conversion events, testing a Meta connection, and retrieving usage statistics. The server calls the server-track-lab API using the configured API key.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"surface-ops-server-track-lab": {
"command": "npx",
"args": [
"-y",
"server-track-lab-mcp"
],
"env": {
"SERVER_TRACK_LAB_API_KEY": "<SERVER_TRACK_LAB_API_KEY>"
}
}
}
}
Run commandnpx -y server-track-lab-mcp
Environment variables to set
SERVER_TRACK_LAB_API_KEYSecretYour server-track-lab API key

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmserver-track-lab-mcp0.1.0

Tools

3

Saved tool list · updated 4d ago

ParameterTypeRequiredDescription
testbooleanNoMark as a test event
userobjectYesAt least one identifier (email, phone, external_id, fbp/fbc, gclid...)
user.ipstringNo
user.fbcstringNo
user.fbpstringNo
user.emailstringNo
user.gclidstringNo
user.phonestringNo
user.countrystringNo
user.li_fat_idstringNo
user.user_agentstringNo
user.external_idstringNo
valuenumberNo
currencystringNoISO 4217, required with value
event_idstringYesStable unique id for this conversion (also used by Meta for Pixel dedup)
timestampstringNoISO 8601; defaults to now
event_namestringYese.g. lead, sign_up, purchase, subscribe, demo_booked
destinationsarrayNo
event_source_urlstringNo

Details

From the project README

server-track-lab MCP server

Let AI agents send server-side conversion events (leads, signups, purchases, subscriptions) to ad platforms through server-track-lab, using your own server-track-lab API key.

Tools

ToolWhat it does
track_eventSend one conversion. server-track-lab normalizes, SHA-256 hashes and dedupes it (by event_id), then forwards it.
test_meta_connectionSend one synthetic Lead to Meta Events Manager → Test events with your saved pixel, token and test code.
get_statsThis month's accepted / deduplicated / failed events, free allowance, and the human vs agent usage split.

Honest status (Oct 2026)

  • Meta Conversions API: live, with your own pixel ID and access token.
  • Google: Data Manager API validate-only preview. Google checks the request; nothing is recorded.
  • LinkedIn: dry-run. The payload is built and shown, not sent.
  • Pricing: first 1,000 billed events each month free, then CA$2 per 1,000. Agent traffic is welcome, counted in its own bucket and billed the same as human traffic.

Setup

  1. Get an API key at https://server-track-lab.pages.dev/signup (a test key sst_test_… never counts toward usage).
  2. Add to your MCP client config:
{
  "mcpServers": {
    "server-track-lab": {
      "command": "npx",
      "args": ["-y", "server-track-lab-mcp"],
      "env": { "SERVER_TRACK_LAB_API_KEY": "sst_live_or_test_key" }
    }
  }
}

Optional: SERVER_TRACK_LAB_BASE_URL (defaults to https://server-track-lab.pages.dev).

The server talks only to the server-track-lab API and sends the header X-Client-Type: agent. Docs: https://server-track-lab.pages.dev/docs

MIT licensed · by Canada Site Ops

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.