The full inventory across discovery, monitoring, alerting, AI, security and developer tools. Things in flight are marked Coming Soon. Things we don't intend to build aren't on this page.
Point us at a host. We find every service, container, domain, cron and log path attached to it.
Services, containers, certificates, cron jobs, network interfaces — all from one agent install.
→Domains pulled from web-server config. Manual add for properties hosted elsewhere.
→systemd services + listening ports auto-detected. No hardcoded list.
→Hetzner, DigitalOcean, Vultr, Cloudflare and more. Bring servers and DNS in via API.
→Live connection graph. Cross-host dependency map.
→Tracked config files, change diffs, drift detection across the fleet.
→key:value tags + dynamic metadata for the parts of your fleet that have to be modeled.
→Geo-resolved hosts on a real map. Status-coloured pins. Manual override.
→Color-coded grouping with aggregate stats per group.
→Every metric, every service plugin, every log — one timeline.
CPU, memory, disk, load, network. Sparklines in lists, full charts on detail.
→MySQL · PostgreSQL · Redis · Nginx · Docker — discovered, parsed, charted.
→HTTP/HTTPS checks per domain — root + sub-checks for paths and APIs.
→Cron + job pings via slug endpoints. Late + dead alerts when the cron stalls.
→Live tail across every host. Auto-discovers paths from web-server config.
→Auto-discovery, per-container metrics, remote start / stop / restart.
→Laravel + Redis + database queues. Wait, throughput, failures.
→Composable widget grid. Per-team views, scheduled reports.
→Disk forecasts based on usage trend. Days-until-threshold per volume.
→Scheduled exports for fleet, monitors, incidents, costs. CSV + PDF.
→Every HTTP request above the threshold — route, duration, top-3 SQL queries. Platform self-monitors with it.
→Toggle slow-log, tail it inline, daily top-offenders digest.
→pg_stat_statements top-N. Parameterised queries, sortable by mean / max / total time.
→Sessions stuck on locks > 5 s. The class of slowness CPU alerts miss.
→On-demand CPU forensics — agent samples top + DB activity when CPU > threshold, auto-stops, AI analyses.
→Pre-built, branded. Reflects monitors automatically.
→Signal over noise. Auto-recovery before the page even fires.
Multi-source rules with include/exclude targets. Anti-flapping, durations, severities.
→Restart the service before paging the human. Per-service policies + cooldowns.
→Statistical baselines per metric. Z-score scored, severity classified.
→Auto-created on alert. Logs from 5 min before crash automatically attached.
→Incident timeline → postmortem. Root cause, impact, follow-ups.
→One-off or recurring. Per-server or fleet-wide. Pause toggle, no edit dance.
→Slack, Discord, email, webhook, push. Routed per rule, per severity.
→Bell + archive. Inline action buttons. Per-event opt-out.
→AI that reads your fleet. Automations that handle the tedious parts.
Incidents summarised, ranked, deduped — across logs, metrics and recent changes.
→Ask why the spike. Get the answer with the evidence — logs, metrics, deploys correlated.
→Time-travel through fleet state right before, during and after an incident.
→Slide-in chat that knows the page you're on. Ask, scope, refine.
→Reusable command playbooks. Run on alert, on schedule, or on demand.
→Triggers × playbooks. Restart, page, ticket, webhook, run a recipe.
→Claude · Cursor · Windsurf · ChatGPT — read & write your infra over MCP.
→Attack-grade response, encrypted off-site backups, audit-ready trails.
Live response: 5 s telemetry, IP bans, Cloudflare, AI postmortem.
→Just-in-time SSH via short-lived CA-signed certs with optional approval gate.
→Every gated session optionally captured as an asciinema cast — replayable, exportable.
→Entropy + magic-byte detection on every backup. Blocks encrypted uploads at the off-site gate.
→Client-side AES-256-GCM encrypted off-site copies. Zero-knowledge restore.
→Every action, every API call, every override — searchable, exportable, signed.
→Shadow-AI detection, AI spend tracker, AI Bill of Materials, MCP fleet monitoring.
→Real-time event stream to Splunk, Sentinel, Datadog or Elastic.
→Continuous compliance checks. Built-in controls + custom rules.
→Hide a host from the AI gateway. The toggle itself is web-auth-only — the AI can't flip it back.
→CVE scanning per host (OS packages first, then optional container images).
→HMAC-signed audit entries with verification UI for SOC 2 / ISO 27001 evidence.
→Everything in the UI is scriptable.
40+ commands for managing infrastructure from the terminal.
→22+ tools. Claude / Cursor / Windsurf / ChatGPT integration.
→Every UI action available as an authenticated endpoint.
→Multi-tab, SFTP, port forwarding, key-synced across devices.
→Every event, where you need it. Per-tenant signed.
→⌘K everywhere. Search hosts, jump to pages, run actions.
→Sentinel, Portwatch, Atlasmap, Dockertop — free standalone agents you can use without the platform.
→Every capability above is on by default. None of it requires a sales call to turn on.