# Instant checks vs monitors

HostTracker gives you two different ways to check a site: an **instant check** you run once, right now, and a
**monitor** that keeps checking on a schedule forever. They use the same underlying check engine, but they
answer different questions.

## Instant check: "is it okay right now?"

An instant check is a single, on-demand run of a check against a target - no schedule, no ongoing history, no
alerts. You get one result, immediately, and that's it. It's the right tool when you just want an answer to "is
this working *right now*" without committing to long-term monitoring.

Instant checks are available two ways:

- **Signed in**, from the **IC** area of the app - run one against any target and see the result plus a short
  history of your recent instant checks.
- **Anonymously**, as HostTracker's **free tools** - no account needed. These cover:
  - Website / HTTP check
  - Page speed test
  - Site crawl
  - Ping test
  - Port check
  - Traceroute
  - DNS query
  - Whois lookup
  - SSL certificate check
  - DNSBL (blacklist) check
  - Web Risk (Google Safe Browsing) check

:::note[Anonymous checks are rate-limited]
Without signing in, a browser may run up to **10 checks per day**, and up to **30 checks per day from the
same IP address** (a shared office or carrier connection counts as one IP). Once you hit the limit you'll
see a sign-up prompt - a free account raises the limit substantially. The limit resets at UTC midnight.
:::

## Monitor: "keep telling me if it stops being okay"

A [monitor](/monitors/what-a-monitor-is/) is the same kind of check, but scheduled to run repeatedly at an
[interval](/monitors/intervals/) you choose, from [locations](/monitors/default-locations/) you choose, with
[alert contacts](/alerts/overview/) subscribed to it. Every run is recorded, state changes become **incidents**,
and a confirmed failure triggers an alert. This is the tool for anything you actually depend on staying up.

## Turning an instant check into a monitor

If an instant check's result looks like something you'd want to keep an eye on, you don't have to re-enter the
target from scratch. Every instant-check result carries a **"Turn into a monitor"** action that pre-fills a new
monitor with the same target and check type - you just pick an interval, locations and who to alert, then save.

:::tip[When to reach for which]
Use an instant check to debug a problem happening *right now*, to sanity-check a URL before committing to
monitoring it, or to answer a one-off question ("is this domain blacklisted?"). Use a monitor for anything you
need to know about *the next time it breaks*, not just this once.
:::

## Do it with the API or MCP

A signed-in instant check is `POST /check` (v2, `check:write` scope) with a `type` and `url`; the answer is
the check result inline (no polling). An MCP client uses the `run_instant_check` tool the same way. The
anonymous free tools use a separate, unauthenticated endpoint and share the daily limit described above.

## Related

- [What a monitor is](/monitors/what-a-monitor-is/)
- [Quickstart: your first monitor](/getting-started/quickstart/)
- [Monitor types](/monitors/monitor-types/)