Skip to content

Outage Status Broadcaster

Example prompt: "When our uptime monitor reports downtime, use AI to draft a short incident summary, post it to #incidents on Slack, and email affected customers from our notification list in Google Sheets. Update the message when the service recovers."

The Problem

When a service goes down, the first 10 minutes are chaos. Engineers scramble to diagnose the issue while customer-facing teams field questions with no information. Someone has to write a status update, post it to the internal channel, email affected customers, and then remember to send a follow-up when things are back. In practice, the status update gets written late, the email goes out even later, and half the team finds out about the outage from a customer complaint rather than an internal alert.

How GloriaMundo Solves It

We build a workflow triggered by a downtime alert from your monitoring tool. An LLM step drafts a concise incident summary based on the alert details — which service is affected, when it started, and what the team is investigating. A conditional step checks the severity: for critical outages it broadcasts immediately to both internal and external channels; for minor degradations it posts to Slack only. The workflow sends the drafted message to your #incidents Slack channel for the team, and emails affected customers using a contact list from Google Sheets. When the monitoring tool reports recovery, a second trigger fires, the LLM drafts a resolution message, and the workflow sends follow-up notifications to the same channels. Glass Box preview shows you the drafted messages and recipient list before anything is sent.

Example Workflow Steps

  1. Trigger (webhook): Fires when the uptime monitor detects a service outage or recovery.
  2. Step 1 (conditional): Determine whether this is a new outage or a recovery event.
  3. Step 2 (LLM): Draft an incident summary (for outages) or a resolution message (for recovery), using the alert details — affected service, start time, and current status.
  4. Step 3 (conditional): Check severity. Critical outages proceed to both Slack and email; minor degradations go to Slack only.
  5. Step 4 (integration): Post the status message to #incidents on Slack with severity, affected service, and estimated next update time.
  6. Step 5 (integration): For critical outages, read the affected-customers contact list from Google Sheets and send a status email via Gmail to each contact.

Integrations Used

  • Slack — internal incident channel for real-time team updates
  • Gmail — external status notifications to affected customers
  • Google Sheets — source of the customer notification contact list

Who This Is For

Operations teams, SREs, and engineering leads at companies where downtime directly affects customers and clear, fast communication is expected — SaaS companies, e-commerce platforms, and any service with uptime commitments.

Time & Cost Saved

During an incident, writing and distributing a status update typically takes 10-15 minutes — time better spent on diagnosis and recovery. The follow-up resolution message often gets forgotten entirely or arrives hours late. This workflow handles both within seconds of the monitoring alert, ensuring consistent communication without pulling anyone away from the technical response. Credit costs are low: one LLM step per incident plus a few integration calls.