AI root-cause analysis, self-healing, monitoring, APM, patching and on-call: the tools you'd normally buy from six vendors, now in one product with a single login and bill.
Via the agent or over SSH. Safety gates stop restart loops, and every recovery lands in the auto-heal feed.
Bring your own provider. Separate models for analysis and chat, optional hard cap, per-call cost log.
Answers, not just alerts.
Every incident gets an automatic write-up that reads your real server logs and names the likely cause. Covers services, Docker, pm2, systemd, websites and sustained CPU/memory spikes.
When a service or site goes down, the full root cause lands in your inbox, with one-click approval for any suggested fix.
Ask questions about an incident in plain language and get answers grounded in your own telemetry.
OpenAI, Anthropic Claude, Google Gemini, Groq or self-hosted Ollama. Free options available, and analysis and chat can use different models.
Per-account token limits, alerts at 80% and 100%, an optional hard cap, and a live meter showing what each analysis cost.
Track your own app's LLM spend: ingestion, pricing, anomaly detection and cost-saving recommendations.
Restart what broke before anyone gets paged.
Downed services restarted automatically across systemd, pm2 and Docker, via the agent or over SSH, with safety gates against restart loops.
Container health and crash-loop detection, automatic pm2 app discovery, and a full systemd service grid with heal approval.
Every automatic recovery across all engines, in one timeline.
Apply a complete monitoring setup in one click at install time, including a ready-made Web Server / Nginx template.
A live status grid of which monitoring systems are operational, degraded or down.
One-line installer for Linux, Windows, Docker, Kubernetes. CPU, memory, disk per-mount, network per-NIC, TCP states, top processes — every 60 seconds.
See outages the way your users do.
Ping, TCP-port, HTTP/HTTPS, DNS, SSL expiry. Checks run from multiple regions in the US and Europe. SLA computation and status pages included.
Track how fast your pages load and catch performance regressions over time.
Identify bot traffic hitting your sites, so you can tell real users from crawlers and abuse.
Alerts well before a certificate or domain is due to expire, so nothing lapses by surprise.
APM, RUM and errors in one place.
Six languages. Distributed tracing, request waterfalls, real-user browser performance. ClickHouse-backed — sub-second queries on billions of events.
Errors from sites and apps grouped automatically, with views for what's new, what's recurring and what's getting worse.
Pod-crash detection with AI-suggested fixes, plus opt-in down and recovery emails for services and clusters.
Safe, audited changes to your fleet.
Every incident classified by an LLM, matched to a playbook, executed automatically. Per-asset autonomy controls. Full audit trail of what AI touched and why.
OS & package CVE scanning, risk-scored prioritization, approval flows, maintenance windows. Auto-rollback on health failure.
Continuous control evaluation. Evidence auto-collected at scan time. Auditor-ready PDF exports to support your own compliance work.
Run hardened shell playbooks on any agent. Slot allow-listed (no shell injection ever), cryptographically signed by the backend.
Built for teams that get audited.
Servers, containers, databases, sites, network devices — discovered or registered, all in one place with tags & environments.
One platform, multiple customer tenants. Per-tenant DBs, isolation by default, SaaS-admin console for operators.
Per-agent client certificates issued by our internal CA. SAN-pinned to agent identity. Cert theft can't impersonate a different agent.
Reach the right person, wherever they are.
Alerts go to whoever is on call and escalate automatically if not acknowledged. Tiered rotations, Slack, Teams & PagerDuty, SLA-aware routing.
Install the app on any phone, log in, and get push notifications for every alert on your account. No app store needed.
Incident and recovery notifications on WhatsApp, with deep links straight to the incident.
Send alert emails through your own mail server, so they arrive from your domain and branding.
A welcome tour and setup checklist get new users to their first alert fast.
In progress, not available yet.
Switch between multiple accounts in the mobile app.
Platform error reporting for account admins.