Catch every exception. Skip the SDK.

Metricly is a fully managed Sentry alternative. Send errors straight to a simple HTTP API authenticated by a project key — then group, triage, and resolve them in a fast dashboard.

Start tracking free See the API

Report from anywhere — curl

POST /api/ingest

curl -X POST https://your-metricly.app/api/ingest \
  -H "Content-Type: application/json" \
  -H "X-Metricly-Key: mtr_your_project_key" \
  -d '{
    "level": "error",
    "exception": {
      "type": "TypeError",
      "value": "Cannot read properties of undefined (reading \"id\")",
      "stacktrace": "at handler (/app/api/checkout.ts:42:18)"
    },
    "environment": "production",
    "release": "web@1.4.2",
    "user": { "id": "u_8123" },
    "tags": { "route": "/checkout" }
  }'

Or from your code — fetch

JavaScript

// Drop this in your global error handler. No SDK, no agent.
async function reportError(err) {
  await fetch("https://your-metricly.app/api/ingest", {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "X-Metricly-Key": process.env.METRICLY_KEY,
    },
    body: JSON.stringify({
      level: "error",
      exception: { type: err.name, value: err.message, stacktrace: err.stack },
      environment: process.env.NODE_ENV,
    }),
  });
}

That's the whole integration. No agent, no instrumentation, no version lock-in — just an authenticated HTTP request whenever something breaks.

Read the full API reference →

Everything you need to triage production

No SDK to install

Your app POSTs JSON to one endpoint, authenticated with a project key. Works from any language, runtime, or edge function — even a shell script.

Smart grouping

Events are fingerprinted by exception type and stack signature, so a thousand occurrences collapse into one actionable issue.

Resolve & regress

Mark issues resolved or ignored. If a resolved issue fires again, Metricly reopens it automatically and flags the regression.

Users affected

Attach a user to each event and Metricly tracks distinct impacted users per issue — know whether it hit one person or ten thousand.

Full event context

Stack traces, tags, releases, environments, request URLs and arbitrary extra data — captured verbatim and searchable.

Fully managed

We run the database, scaling, and storage. Create an account, grab a key, and start sending events — zero infrastructure to babysit.

Start catching errors in minutes

Create a project, copy your ingest key, and fire your first event. Free to start — no credit card, no infrastructure to set up.

Create your account