Uptriva MCP

Let your AI agent watch your infrastructure

Uptriva ships a built-in Model Context Protocol server. Connect Claude, Cursor or any MCP-compatible client and your agent gets real monitoring tools — not screenshots of a dashboard.

Start monitoring for free Read the MCP docs

Connect in one line

Point your MCP client at the endpoint with an API token from your dashboard. Works with Claude Desktop, Cursor and every client that speaks MCP.

{
  "mcpServers": {
    "uptriva": {
      "type": "streamable-http",
      "url": "https://api.uptriva.com/api/v1/uptime/mcp",
      "headers": { "Authorization": "Bearer YOUR_API_TOKEN" }
    }
  }
}

Create an API token in the dashboard under Settings → API tokens.

What your agent can do

Monitors

Create, list, update and delete monitors across 38 check types — HTTP, DNS, TCP, databases, game servers, Kubernetes workloads.

Status pages

Create and inspect status pages, their monitors and public configuration.

Incidents

Declare incidents, post status updates and review incident history — your agent can run comms during an outage.

Alerting

Manage notification contacts and escalate through email, SMS, voice, Slack, Telegram and webhooks.

Overview

Workspace-wide state: monitor counts by status, open alerts, recent incidents — the agent's situational awareness.

Full API access

A generic request tool covers the entire Uptriva API surface beyond the dedicated tools, under your token's permissions.

Built for agents that do real ops

Configure while you code

"Watch https://api.myapp.com/health every 15s and page me if it drops" — done, without leaving your editor.

Outage response

Ask what is down, get the failing monitors with latency and error detail, declare the incident and post the status update in one thread.

Scoped by your token

Agents act with your account's permissions. Read-only insights or full control — you decide what the token can touch.