Clicking through a dashboard is fine for ten monitors. For a hundred services, dozens of client sites or infrastructure that changes every week, monitoring has to live in code. The REST API covers monitors, maintenance windows and status updates, documented at app.uptimecheck.com/docs/api; an MCP server lets AI assistants do the same with an API key.
Endpoints
Every endpoint lives under https://app.uptimecheck.com/api/v1:
/monitors: list, create, update and delete monitors of every type, pause and resume them, and read their latest checks./monitors/{id}/metrics,/page-speedand/email: server metrics, Lighthouse results and email authentication reports./incidents: list incidents, read the timeline and acknowledge one./maintenance: schedule, change, end or cancel maintenance windows, and list, change or stop recurring ones under/maintenance-schedules. APATCHwith only the fields to change moves a window, and subscribers who were told about it get the new time./status-pages: list your pages, and post, follow up and delete status updates on them, with an impact per affected component./summary,/projectsand/alert-channels: the state of everything at a glance, and the IDs to use when you create monitors.
Create a monitor from your pipeline
Authenticate with an API key in the Authorization header and send JSON:
curl https://app.uptimecheck.com/api/v1/monitors -H "Authorization: Bearer uc_…" -H "Content-Type: application/json" -d '{"type":"http","name":"Checkout","target":"https://northwind.shop/checkout","intervalSeconds":60}'
Run it in the same step that deploys the service, and every new endpoint is monitored the moment it goes live. The first check runs within a minute. The full reference, with request and response examples, is in the API documentation.
AI assistants over MCP
The MCP server at https://app.uptimecheck.com/api/mcp lets Claude, Cursor, Copilot and other assistants work with your monitoring using the same API keys. Ask for a status summary, look into an incident, test a request, create a monitor, pause one while you deploy, start a maintenance window or post a status update. Tools that change something need a read and write key.
Common uses
Agencies
Onboard a new client in one script: add monitors for the site, the checkout and the API in the client’s project, with the right alert channels. Offboarding is the same script in reverse.
Platform teams
Keep monitors in sync with your service catalog, and start a maintenance window from your deploy script so planned restarts do not open incidents or cost uptime. A Terraform provider for monitors as code is coming soon.
Status communication
Post a status update from your own incident tooling, so the status page and its subscribers hear about it the moment your team does.
Webhooks for the other direction
The API lets you manage Uptimecheck. Webhooks let Uptimecheck call you: every outage and recovery is posted as signed JSON to your own endpoint, ready to connect to whatever runs your on-call process. See alerts and notifications.
Practical tips
- Store the API key as a secret in your CI system, never in the repository.
- Use a read-only key for dashboards and reporting, and a read and write key only where something needs to change.
- Make scripts idempotent: look up a monitor by name or target before creating it, so a re-run does not create duplicates.
- Use the same naming scheme in the API as in your infrastructure, so a monitor is easy to trace back to the service it watches.
Most teams start by scripting the boring part, adding monitors for new client sites or services, and expand from there to maintenance windows and status updates.
API and MCP access are included on Professional and Business. Plans start at €9 per month billed yearly or €12 billed monthly, excl. VAT; compare plans on the pricing page, or start a 14-day free trial of Professional without a card. See API monitoring to monitor your own API with Uptimecheck.