MCP Server Documentation — Connect AI Tools to CronAlert

Overview

The CronAlert MCP server lets AI assistants like Claude, Cursor, and Windsurf manage your monitors directly. Create, update, and monitor uptime without leaving your editor.

You'll need an API key to authenticate. Create a free account, then generate a key in Settings > API Keys.

Available Tools

list_monitors List all monitors with status and response times

get_monitor Get detailed info about a specific monitor

create_monitor Create an HTTP, heartbeat, or TCP port monitor — with keyword, regex, or content checks, and a heartbeat's expected interval and grace period

update_monitor Update configuration, content checking mode, heartbeat interval and grace, or pause/resume

delete_monitor Delete a monitor and all its history

get_check_results Get check history, uptime %, and response times

get_monitor_incidents Get incidents for a specific monitor

list_incidents List all active incidents across monitors

list_status_pages List all configured status pages

add_incident_update Post a status update on an active incident

get_incident_updates Get status updates for an incident

import_monitors Import monitors from UptimeRobot, Pingdom, Better Stack, StatusCake, Checkly, Cronitor, Healthchecks.io, and more, or from CSV/JSON

Heartbeat Monitors

Since version 1.5.0, create_monitor and update_monitor accept type: "heartbeat" with an expected interval and an optional grace period. Ask your assistant, for example: "Create a heartbeat monitor called Nightly DB backup that expects a ping every 24 hours with a 2-hour grace period." The tool sends checkInterval: 86400 and gracePeriod: 7200; the response includes the ping URL to add to the end of your job.

checkInterval — seconds between expected pings, from your plan's minimum (60 on Pro and above) up to 604800 (7 days). For HTTP and TCP monitors this field is ignored and set from your plan.
gracePeriod — extra seconds a ping may be late before the monitor is marked down. Defaults to the interval, capped at one hour, so a 1-minute heartbeat alerts after 2 minutes and a daily one after 25 hours.

Heartbeat and TCP monitors require the Pro plan or above.

Installation

Claude Code

claude mcp add cronalert -e CRONALERT_API_KEY=YOUR_API_KEY -- npx -y cronalert-mcp

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "cronalert": {
      "command": "npx",
      "args": ["-y", "cronalert-mcp"],
      "env": { "CRONALERT_API_KEY": "YOUR_API_KEY" }
    }
  }
}

Cursor

Add to .cursor/mcp.json in your project root

{
  "mcpServers": {
    "cronalert": {
      "command": "npx",
      "args": ["-y", "cronalert-mcp"],
      "env": { "CRONALERT_API_KEY": "YOUR_API_KEY" }
    }
  }
}

Windsurf

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "cronalert": {
      "command": "npx",
      "args": ["-y", "cronalert-mcp"],
      "env": { "CRONALERT_API_KEY": "YOUR_API_KEY" }
    }
  }
}

Remote Server

Connect any MCP client directly to our hosted endpoint — no local install needed.

Endpoint https://cronalert.com/mcp

Auth Authorization: Bearer YOUR_API_KEY

Transport Streamable HTTP

Agent Skill

The MCP server gives your agent the tools; the open-source cronalert-uptime-monitoring skill tells it when and how to use them: which monitor type fits a URL, a cron job, or a bare IP, how to pick a heartbeat grace period, how to triage an outage, and what to confirm before deleting. Copy skills/cronalert-uptime-monitoring from the GitHub repo into .claude/skills/ (Claude Code) or ~/.codex/skills/ (Codex), or install it from Agensi. Read the guide.

Resources

Current release: cronalert-mcp@1.5.1, also listed in the official MCP registry.

npm GitHub Smithery Glama Agensi skill

Setup Guides

For detailed walkthroughs with practical examples, see our blog posts: