MCP server for uptime and status

Quick answer

StillOnline MCP (npx -y stillonline-mcp) exposes your monitoring data to AI agents: list projects, read check status, and manage incidents from Cursor, Claude Code, or other MCP clients. Requires Pro ($9/mo) or Ultimate ($29/mo) and an API key from settings. Free does not include API or MCP access.

Benefits

  • Post-deploy checks — ask an agent to verify production after a merge.
  • Same data as REST — MCP tools wrap the private v1 API.
  • No custom scraper — agents use structured tools instead of parsing HTML.
  • Public status JSON — combine with /v1/public/status/{slug} for read-only agent workflows.
  • npm distribution — install without cloning a repo.

How it works

  1. Upgrade to Pro or Ultimate.
  2. Create an API key in dashboard settings.
  3. Add MCP config to Cursor or Claude (see MCP docs).
  4. Run npx -y stillonline-mcp with your key.
  5. Agent tools: projects, checks, incidents (per OpenAPI).

Full setup steps: docs/mcp.

Plan availability

PlanMCPAPI key
FreeNoNo
ProYesYes
UltimateYesYes

FAQ

Does UptimeRobot offer MCP?

UptimeRobot does not ship an official MCP server for agent workflows. StillOnline targets indie SaaS + AI agents with stillonline-mcp on npm.

Which MCP clients work?

Any client that supports stdio MCP servers—Cursor, Claude Desktop, Claude Code, and similar. Configuration examples are in docs/mcp.

Can MCP create checks without the dashboard?

Yes on paid plans—agents can use API-backed tools per your key permissions. Start with read-only workflows if you prefer.

Is MCP included on the free plan?

No. Free has no REST API or MCP. Upgrade to Pro for API + MCP.