Alerts and usage warnings

Hear about findings, tripwire hits, silent sources, and incidents by email, Slack, Discord, Teams, or any URL, and about your week every Monday.

Core Available on Email on all plans; Slack, Discord, Teams, and custom URL alerts from Starter Updated 25 October 2026

These pages describe ShellOrbit as it runs today. Breaking changes to the HTTP API are announced before they ship, and existing request and response shapes stay supported.

When alerts are sent

Alerts and usage warnings: What happened, When you hear
What happenedWhen you hear
A new critical or high finding from a source ruleStraight away, once per finding; repeats update it without alerting again
A source goes silentStraight away, once until events resume
A new critical or high finding from a site checkStraight away, at most once an hour for each domain
A new critical or high finding from a log drain or the SDK, including a new outbound hostStraight away, at most once an hour for each drain or project
A tripwire is hitWithin seconds, then quiet for 15 minutes for that tripwire
A critical or high incident affects your stackStraight away, at most once in 24 hours for each incident
A revoke is scheduled, cancelled, done, or failsStraight away
A log drain reaches the month’s log allowanceOnce that month
80 percent of the month’s fair use source events are usedOnce that month. See limits
A free trial ends in about two daysOnce, with the date and first payment

Channels

Alerts and usage warnings: Channel, Plans, Setup
ChannelPlansSetup
EmailAllOwners and admins, plus extra addresses on the alerts page
SlackStarter, Pro, BusinessPaste an incoming webhook URL (https://hooks.slack.com/services/...)
DiscordStarter, Pro, BusinessChannel settings, Integrations, Webhooks; paste the URL (https://discord.com/api/webhooks/...). Mentions are off
Teams or any HTTPS URLStarter, Pro, BusinessTeams, Mattermost, or your own handler; gets the JSON below
json
{
  "text": "ShellOrbit: critical finding \"CloudTrail logging stopped\" from the source \"AWS production\". ...",
  "content": "...the same text...",
  "message": "...the same text...",
  "url": "https://app.shellorbit.com/..."
}

Alerts never contain source event bodies or keys. A channel URL is a credential: it is shown once, then masked. Private addresses are refused and redirects are not followed.

Test your alerts

Save your settings, then press Send test alert or call POST /v1/alerts/test. Each channel reports sent, or why not.

Weekly digest

Every Monday around 08:00 UTC, to the same addresses, covering Monday to Sunday:

  • New findings by severity and by source, and how many were resolved.
  • Sources that went silent, and incidents affecting you.
  • Tripwire hits, and this month’s use of source events.

Quiet weeks send nothing. On by default on every plan; turn it off on the alerts page.