If our database is briefly unavailable (for example during maintenance), the API now answers 503 with the error code service_unavailable and a Retry-After header instead of a generic error, and recovers on its own. Retry after the number of seconds it gives.
Security updates across the web app and the API, including the latest framework and dependency releases.
new
Stack watch: security advisories for the software you run
List the software you run, optionally with its version, and get alerted on your existing alert channels when NVD or CISA's Known Exploited Vulnerabilities catalog publishes an advisory that matches.
Only news is pushed: advisories that predate an item are listed but never alerted, and you choose the threshold (known-exploited only by default).
Watch a whole dependency tree by uploading a lockfile or SBOM (CycloneDX/SPDX, npm, pnpm, Yarn, Composer, pip, Go, Bundler, Cargo), checked against OSV.dev. The upload endpoint works as a CI step.
HTTP monitors now show the software a site discloses (Server, X-Powered-By) with a one-click Watch.
New API endpoints under /v1/stack and the read-only MCP tools list_watched_software and get_software_advisories. Available on Solo and above.
improved
Status page logos, incident history and app.hypermonitors.com
Upload your logo straight to a status page (PNG, JPEG, WebP, GIF or SVG, up to 512 KB) and set its width; wide logos now keep their aspect ratio instead of being squeezed into a square.
Status pages list the five most recent incidents, with “show more” for the rest of the page’s history window.
Sign-in and the dashboard moved to app.hypermonitors.com. Old links redirect, and one login works on both hosts.
improved
Cookie settings
Visitors from the EU, EEA and UK get a cookie banner with equally weighted “Necessary only” and “Accept all” choices, and anyone can change their choice in the cookie settings on the privacy page.
fixed
More accurate public uptime
Status-page uptime bars and badges now use the same cross-region confirmation as incidents, so a single slow region no longer shows as downtime next to “no incidents”.
Public status pages load faster, and a rendering error that could break some pages is fixed.
Subscriber emails now say why the recipient got them, next to the unsubscribe link.
fixed
MCP server fixes
Incident updates and resolution notes posted through MCP now appear on status pages exactly as written; text that had been stored URL-encoded was repaired.
create_monitor now lists the domain-expiry monitor type and rejects intervals a monitor type cannot honour instead of silently changing them.
The MCP tool manifest is public at /v1/mcp/manifest, so you can inspect the tools before connecting.
new
Incident updates, states and resolutions
Incidents now move through investigating → identified → monitoring → resolved, so an incident says how far along you are rather than just open or closed.
Post updates to an incident as you learn more. They appear on any status page showing that monitor, alongside the ones you write by hand.
Resolving an incident asks what fixed it, and that note is kept with the incident and exported in the CSV.
Incident actions moved into the ⋯ menu on each row, matching the monitor list, and rows expand to show the full timeline.
Posting an update also acknowledges the incident, so repeat alerts stop without a separate click.
Updates and resolution notes stay private unless the status page has “show incident details” turned on.
New API endpoints and MCP tools for reading a timeline, posting updates and resolving with a note.
new
SAML single sign-on for Enterprise
Enterprise orgs can connect their identity provider (Okta, Azure AD, Google Workspace…) with a SAML metadata URL.
Teammates on your verified email domains sign in with “Continue with SSO” and are added to your organization automatically on first login.
improved
Smarter regions & in-app support
New monitors default to automatic regions: we check from every region your plan allows and order them by response time, fastest first.
A dedicated Support & feature-request page, reachable from the sidebar.
Plan upgrades now unlock the extra regions on your existing monitors automatically.
improved
Faster dashboard & mobile polish
Rebuilt the monitor list and charts for quicker loads on large fleets.
The mobile bottom bar was tidied up with a “More” menu so the primary destinations stay one tap away.
new
German language support
The whole app, emails, and public status pages are now available in German (Deutsch).
Set your language per account, and an organization default for team-wide emails.
new
Model Context Protocol (MCP) server
Connect HyperMonitors to Claude and other MCP clients to query monitors, incidents, and uptime in natural language.
Manage your monitors from your AI assistant — a first among uptime tools.
new
On-call rotations & escalation
Define on-call schedules and route escalations to whoever is on call right now.
Per-monitor repeat notifications and time-based escalation to extra channels.
new
Audit log & API keys
Team and Enterprise orgs get an audit log of who changed what.
Scoped API keys with per-key rate limits for automation and CI.
new
Domain expiry, HTTP methods, tags & bulk actions
New domain-expiry monitor type — we watch your domain registration via RDAP and remind you before it lapses.
HTTP monitors now support GET/HEAD/POST/PUT with an optional request body, for real API monitoring.
Exposed advanced monitor options: expected status codes, follow-redirects, keyword-absent, timeout, confirmation threshold, and heartbeat grace.
Tag your monitors and filter by tag; select multiple monitors to pause, resume, or delete in bulk.
improved
Growth & docs
A dashboard onboarding checklist guides you from first monitor to a production-ready setup.
Public developer docs at /docs — the full REST API reference plus MCP server setup, no login required.
Optional lifecycle emails, including a weekly uptime digest you can turn off any time.
new
Status pages v2 — subscribers & incident updates
Visitors can subscribe to a status page (double opt-in) and get emailed on incidents and maintenance.
Post manual incidents with an update timeline (investigating → identified → monitoring → resolved).
Embeddable status/uptime SVG badges, and self-serve custom domains that verify automatically.
new
Alerting depth & new channels
Maintenance windows suppress alerts during planned work while still recording data.
Repeat + escalating notifications, incident acknowledgement, and response-time (degraded) alerts.
New alert channels: generic signed webhook, Microsoft Teams, and PagerDuty.
new
Multi-region probes
Every check can run from multiple regions; a single flaky vantage point no longer pages you — outages are confirmed across regions first.
Regions became a plan axis: 1 (Free) / 2 (Solo) / 3 (Team) / all (Enterprise).
improved
Launch essentials
Password reset, change email, account deletion (GDPR), and Google/GitHub sign-in.
Privacy, terms, and imprint pages; full SEO metadata, sitemap, and OpenGraph images.
Real annual-billing discounts across every paid tier.
new
HyperMonitors is live
HTTP, port, ping, and heartbeat monitors with configurable intervals.
Incident detection with a proper up/down state machine, plus email and Telegram alerts.
Public status pages, SSL certificate expiry reminders, and team organizations.