Uptime monitoring API for agents, CI/CD, and automation

Use the KEEPitALIVE API from scripts, CI/CD, internal tools, and autonomous agents to manage operational work with scoped permissions.

What an agent can manage

With the required scopes and within the account plan, an agent can read monitor health and history; create, update, pause, resume, group, share, and delete monitors; manage incidents and comments; configure triggers, notification resources, status pages, embeds, and API keys. The API is versioned and documented with an OpenAPI specification.

Use the API safely

Create a scoped API key for the specific automation instead of sharing a broad personal credential. Fetch the current state before a mutation, respect validation and rate limits, and use documented idempotency controls where available. Require explicit human approval for destructive changes, credential rotation, billing, and account deletion.

curl -H "Authorization: Bearer $KEEPITALIVE_API_KEY" \
  https://keepitalive.dev/api/v1/monitors

Connect an AI agent with MCP

The KEEPitALIVE MCP server lets an MCP client inspect and manage your monitoring through the same scoped API key. It connects to the production API by default and uses write mode when no mode is configured. Set read mode explicitly for a least-privilege starting point; enable write mode when you want an agent to make reversible operational changes.

{
  "mcpServers": {
    "keepitalive": {
      "command": "npx",
      "args": ["-y", "@keepitalive/mcp"],
      "env": {
        "KEEPITALIVE_API_KEY": "kia_live_...",
        "KEEPITALIVE_MCP_MODE": "read"
      }
    }
  }
}

Choose the smallest tool surface

read exposes monitoring and incident data only. write is the default and adds reversible actions such as creating or editing monitors, pausing or resuming checks, and updating incidents. full additionally exposes permanent delete tools. Your API key scopes always remain the hard permission boundary; MCP mode can only narrow them further.

Build an operational loop

A CI job can create or update a monitor after deployment. An agent can investigate incidents using check history, add a comment, pause a known maintenance window, or configure a trigger to notify an existing workflow. Keep automated actions narrow, observable, and reversible.

Use the authoritative contract

The API reference explains endpoints and authentication. OpenAPI is the machine-readable source of truth for tools, and the agent guide covers approval and safety expectations before an agent changes production resources.

Related