Features
Built for on-call engineers, SREs, and engineering managers who run production systems. No roadmap items on this page.
Datadog, New Relic, and generic webhooks, with a PagerDuty inbound payload preset and a dynamic field mapper for anything custom.
Rotations with overrides, a setup wizard, calendar export, and readiness checks that flag who cannot actually be paged.
Per-rung channels and delays over SMS, voice, WhatsApp, email, Slack, and Teams, on your own Twilio and Meta credentials.
iOS critical alerts and Android high-priority push break through Do Not Disturb. Acknowledge in one tap.
Inbound alarms that clear can close the incident they opened, so nothing sits open because the page stopped firing.
Slash commands, dedicated incident channels, two-way message mirroring, and Adaptive Card DMs on both platforms.
Zoom, Google Meet, or Microsoft Teams calls provisioned on declare and posted to the channel automatically.
Versioned runbooks linked to services, with steps that execute HTTP request actions instead of only describing them.
Policy-driven nudges for stale incidents, overdue updates, and unfinished post-mortems, with owner mute and snooze.
Timestamped updates sent to stakeholder groups and mirrored into the channel timeline.
A branded public page on your own domain, with component groups, maintenance windows, and email/webhook subscribers. See how status pages work.
Drafted from the real timeline: events, updates, deploys, runbook runs, and related incidents. A non-AI template path is always available.
Recurring causes cluster with an owning team, a status, severity-weighted burden, and the follow-ups nobody finished.
A graph of related incidents and the deploys behind them, so systemic failure looks like a pattern rather than bad luck.
Action items become Jira or Linear issues, with completion and aging tracked as a metric.
Published post-mortems land in Confluence or Notion in your own template, with the variables already filled in.
An agentic assistant over incidents, alarms, on-call, services, insights, and post-mortems. Global, or scoped to one incident.
Read tools run inside the loop. Mutating tools surface as a pending call with exact arguments and wait for an explicit approval.
api.waroom.co/mcp exposes the full tool registry with its own OAuth authorization server and consent screen, sharing one registry with Ask AI.
Anthropic, OpenAI, or Google keys, with Langfuse tracing, token budgets, and per-org usage metering. No markup.
Scoped API keys, in-app docs, signed outbound webhooks, and a delivery log with signature-failure surfacing.
A GitHub App maps repos to services and teams, including monorepo deploy targets, and links every deploy to the incidents it caused.
