Back to Blog
ai

MCP Is Here: Query Your Monitors from Claude, Cursor & Windsurf

3 min read
March 31, 2026
Morten PradsgaardBy Morten Pradsgaard

Connect exit1.dev to your AI assistant via the Model Context Protocol. Ask about uptime, failures, and response times in plain English.

Free Uptime Monitor

10 free monitors with 5-minute checks. Unlimited monitors and 1-minute checks with Nano. See the feature deep dive and migration checklist.

Analytics & Reports

Turn uptime data into SLAs, client reports, and long-term performance trends without bolting on another BI tool.

Smart Alerting

Ship webhook and email alerts that your team actually respects. Integrate PagerDuty, Opsgenie, Slack, or Discord without paying a premium.

MCP Is Here: Query Your Monitors from AI

Dashboards are great until you're in the middle of something else and just need a quick answer. "Is anything down?" "What's the uptime this month?" "Show me the last few failures." You shouldn't have to open a new tab and click through three screens for that.

That's why we built an MCP integration for exit1.dev.

What Is MCP?

MCP — Model Context Protocol — is an open standard that lets AI assistants talk to external tools. Think of it as a bridge between your AI (Claude, Cursor, Windsurf) and your monitoring data. Once connected, your AI assistant can pull live data from exit1.dev and answer questions about your checks, uptime, and incidents.

No dashboards. No API calls. Just ask.

What You Can Do

Once connected, your AI assistant has access to 5 read-only tools:

  • List checks — see all your monitors and their current status
  • Get check details — response time, SSL info, maintenance windows
  • Pull history — timestamps, status codes, and errors for any check
  • Query stats — uptime %, average/min/max response times across time ranges
  • View status pages — current snapshot of any public status page

Some real prompts you can try:

  • "Are any of my monitors down right now?"
  • "What's the uptime for my API check over the last 30 days?"
  • "Show me the last 10 failures for production"
  • "Compare response times this week vs last week"

Setup in 2 Minutes

  1. Create an API key in your exit1.dev dashboard (Settings > API Keys) with checks:read scope
  2. Add the MCP config to your AI assistant:
{
  "mcpServers": {
    "exit1": {
      "command": "npx",
      "args": ["-y", "exit1-mcp"],
      "env": {
        "EXIT1_API_KEY": "ek_live_your_key_here"
      }
    }
  }
}
  1. Restart your AI assistant — that's it

We have setup instructions for Claude Desktop, Claude Code, Cursor, and Windsurf in the docs.

Why Read-Only?

The MCP server only reads data. It can't create, modify, or delete checks. Your API key only needs checks:read scope. This is intentional — your AI assistant should be able to answer questions, not accidentally delete your production monitors.

Who Gets It?

MCP is available on the Nano plan ($3/month) and above. It joins unlimited monitors, SMS alerts, domain intelligence, maintenance mode, and everything else that makes Nano the plan for professionals.

If you're on the free plan, upgrade to Nano to start using it.

What's Next

This is v1 — read-only access to your monitoring data. We're already thinking about what comes next: write operations, alert management, and deeper integrations with AI workflows. If you have ideas, drop by our Discord and let us know.

Full setup guide: docs.exit1.dev/integrations/mcp

Feature page: exit1.dev/mcp

Morten Pradsgaard is the founder of exit1.dev — the free uptime monitor for people who actually ship. He writes no-bullshit guides on monitoring, reliability, and building software that doesn't crumble under pressure.

Start Monitoring