RealUptime

npx add-mcp 'https://mcp.realuptime.io/mcp'

What it exposes, honestly

The keyless endpoint (https://mcp.realuptime.io/public) serves exactly five read-only tools, all over RealUptime's own probe measurements of the services on the outage tracker. Nothing here reads or changes an account: no monitors, no status pages, no incidents. That is what the separate, key-authenticated endpoint does, documented in full on the technical reference.

Every answer carries the regions actually read, a measurement timestamp, an explicit freshness verdict, and a URL to cite. "Blocked" means our probe was refused and is never reported as "down"; a no-data answer is never a healthy answer; a service we do not track answers "not covered" rather than a guess. Same rules as the rest of the tracker, set out in full on the methodology page.

Claude Code

claude mcp add --transport http realuptime-outages https://mcp.realuptime.io/public

Claude Desktop, or any MCP client using a JSON config

{
  "mcpServers": {
    "realuptime-outages": {
      "url": "https://mcp.realuptime.io/public"
    }
  }
}

Where to find it

• Smithery
• Glama
• Official MCP registry (io.realuptime/mcp)

Limits

60 requests per minute per IP address: an anonymous call cannot be attributed to an account, and unlike the public outage JSON endpoint an MCP response cannot be cached, so every call is a real read. One assistant question costs about four requests, roughly fifteen questions a minute from one address. If you share an outbound address with a lot of other people, or need sustained access, the key-authenticated endpoint has its own larger budget; see the full reference for setup.

Your own monitors, from the same server

The same MCP server also exposes your RealUptime account over a second, key-authenticated endpoint: list and manage monitors, read status pages, and open or update incidents, mirroring the REST API exactly, on every plan. Plan limits apply as in the dashboard. Full tool list, authentication, and setup on the MCP server reference.