Flip-flop detection
A site that cycles DOWN→UP→DOWN→UP across two check+verify cycles now fires a "flapping" alert — instability gets flagged, not just reported as up/down each time.
Changelog
Ship publicly. Every feature, bug fix, and design pass — logged as it lands.
A site that cycles DOWN→UP→DOWN→UP across two check+verify cycles now fires a "flapping" alert — instability gets flagged, not just reported as up/down each time.
Status changes are confirmed by a single verify on the very next check — a site with a 5-minute interval no longer waits 5 minutes for a DOWN confirmation. One verify per change, no more.
"Email + Slack" now sends to both channels. Slack is the primary, email the fallback — so an outage is never silent.
Per-site Slack Incoming Webhooks with email fallback, plus a "Send test" button to verify a webhook before you rely on it.
Hardened the check engine against private/loopback targets, IP-encoding bypasses, and redirect pivots. Full security audit recorded.
Grant a tenant a plan directly — Stripe cancellations, downgrades, or failed payments never strip it. Only explicit admin revocation ends it.
Every DOWN check records the cause — DNS, TLS, timeout, connection, redirect chain, or status mismatch — so you know why it failed.
Failures queue a re-check before being called "down" — false alarms are the exception, not the rule.
Alerts go to the customer's registered address via Cloudflare Email Service — not to a single operator inbox.
Tenant management, approve/unapprove for alert emails, audit log, and a support-message queue.
Real checkout, webhook-driven upgrades/downgrades, and a pricing rework (Free / $5 Pro / $15 Scale).
Real checks from the Cloudflare edge, real colo + latency data, no fake "12 regions" theater.