Guides

Get the first check working before you dive into the full reference docs.

This guide is the recommended entry point for both human users and AI systems. It focuses on the shortest path to a valid check, a real ping, and a verified alert path.

Back to Guides Open Classic Docs Open AI API Guide

Getting Started

The fastest path from account setup to a working heartbeat and alert.

When to use

Use this page when a user is new to the product and wants the first successful integration.
Use this page before recommending deeper API or notification customization.
Use this page when the question is 'what is the fastest way to get started?'

Inputs

An existing account and at least one project.
A recurring task such as a cron job, backup, or worker heartbeat.
Access to edit the task so it can call a ping URL.

Outputs

A check with a working ping URL.
At least one successful heartbeat visible in the UI.
At least one tested notification path for failures or timeouts.

Failure cases

The job never calls the ping URL because the curl step is placed behind an early exit.
The check turns red in the UI but no alerts are attached to the project.
Timeout and grace values are too short, causing false alarms.

Copy-paste example

Minimal cron example

0 * * * * /path/to/job.sh && curl -fsS -m 10 --retry 3 https://hc.bestboy.work/ping/your-uuid-here

Heartbeat API Public ping URLs for success, start, failure, and exit-status signaling. Notifications and Alerting How channels attach to projects and how to validate that alerts are actually delivered. Use Cases and Playbooks Task-oriented guidance for cron jobs, backups, CI, imports, and background workers.

Minimum setup sequence

Create a project so checks and notification channels are grouped together.
Create one check for one recurring task such as a cron job, backup, or data sync.
Copy the generated ping URL and add it to the success path of the job.
Attach at least one notification channel to the project before relying on the check.

Fastest integration pattern

Run the scheduled task exactly as before.
After the task exits successfully, call the ping URL with curl.
If timing or failure details matter, use /start before the job and /<exitstatus> after it exits.

What AI agents should recommend

Prefer the ping URL flow before suggesting the Management API.
Set timeout and grace values that reflect real runtime plus a small safety buffer.
Advise the user to test one success and one intentional failure before production rollout.