Know your site is down before your customers do.

CheckHTTP requests your site every minute from runners on AWS, Google Cloud and Azure.

  • Free plan, no credit card
  • 60-second checks
Monitors Live 3 up1 degraded
shop.example.comHTTPS · every 60s
99.97%142 ms
api.example.com/healthKeyword "ok" · every 60s
100%88 ms
Checkout flowScripted check · every 5 min
99.4%1.9 s
status.example.comHTTPS · SSL valid 74 days
100%61 ms
Response time · shop.example.com · last hourp50 128 ms
Resolved
shop.example.com returned 502 for 3 minutesConfirmed from aws/us-east-1 and gcp/us-west-1 · HAR attached · email + webhook sent
Uptime & keyword checksScripted browser checksReal-user monitoringSSL expiryPublic status pagesMCP for AI agents

Every failure comes with evidence.

Each check records the HTTP status, DNS, connect, TLS and time-to-first-byte, which runner made it and a HAR file you can open in your browser's dev tools. You see why it failed, not just that it did.

See all check types →
CheckHTTP site page: 24-hour uptime strip, 60 s check frequency, last runner, average latency and a DNS, connect, TLS and TTFB breakdown

Alerts that wait for a second opinion.

Choose how many runners must fail, for how many checks in a row, before anyone is woken up. One flaky network path won't page you; a real outage will, by email or webhook.

Set up alerts →
Recent checks table: time, status, SSL, HTTP status, latency, runner and HAR link for each check

See what real visitors experience.

Add a small script and CheckHTTP collects Core Web Vitals (LCP, INP, CLS), slow pages and JavaScript errors from real browsers, next to your uptime checks.

How real-user monitoring works →
Real-user monitoring dashboard: traffic over 24 hours, pageviews, errors, TTFB and Core Web Vitals

Checked from more than one cloud.

Runners on AWS, Google Cloud and Azure check your site independently, so a single provider's bad day doesn't look like your outage. This map shows the runners online right now.

Cloud status directory →
Runner networklive
Locating runners…
AWSGCPAzure

Let your AI agent manage monitors.

CheckHTTP has an MCP server. Add it to Claude, Cursor or any MCP client and your agent can create monitors, read check results and investigate failures for you.

MCP setup guide →
mcp.jsonMCP · SSE
{
  "mcpServers": {
    "checkhttp": {
      "url": "https://checkhttp.com/mcp/sse",
      "type": "sse",
      "headers": { "Authorization": "Bearer YOUR_API_KEY" }
    }
  }
}

Simple pricing

Free

$0

  • 1 site
  • 60-second checks from US West
  • Email alerts
  • 4k real-user events / month
Start free

Professional

$15/mo

  • 10 sites
  • Global runners
  • 1M real-user events / month
  • Priority support
Choose Professional

Compare plans in detail →

Start watching your site in a minute.

Paste a URL. Your first checks run immediately.

Add your site

Illustrations by Streamline (CC BY 4.0)