Outgoing Webhooks

Outgoing webhooks let Rela AI notify external systems automatically when something happens — a critical task is created, an alarm is acknowledged, a maintenance plan executes. Instead of polling the system for updates, the external system receives a push notification at the exact moment the event occurs.

What is it for?

Webhooks solve the integration problem for systems that cannot be directly connected through the marketplace. Any external system that can receive an HTTP request can be connected via webhook:

• Slack or Microsoft Teams: receive a message in a channel when a critical alarm fires
• CMMS or ERP: automatically create a work order in the external system when Rela AI creates a task
• Custom monitoring dashboards: push real-time event data to an internal visualization tool
• Incident management systems: create incidents in PagerDuty or Opsgenie when critical events trigger

How does it work?

You configure a webhook with the URL of the external system and the events you want to forward. When one of those events occurs in Rela AI, the system immediately sends the event data to that URL. If the delivery fails, the system automatically retries up to 5 times with increasing wait times.

How to use it?

Create a webhook

1. Go to Integrations > Webhooks in the sidebar.
2. Click New Webhook.
3. Configure:
   • Name: a descriptive label for this webhook (e.g., "Critical Alarms to Slack")
   • Destination URL: the URL of the external system that will receive the notifications
   • Events: which events trigger this webhook (alarm created, task created, maintenance executed, etc.)
   • Filters: optional conditions to only forward specific events (e.g., only Critical or Emergency alarms)
   • Authentication: if the external system requires an API key or authorization header, add it here
4. Save.

Test before activating

Before activating the webhook in production, run a test:

1. Select the webhook.
2. Click Test.
3. The system sends a sample notification to the destination URL and shows the result:
   • Whether the external system responded successfully
   • What response it returned
   • How long the response took

This verifies the URL is correct, the external system is reachable, and the authentication headers work — all without generating any real events.

Monitor delivery health

Each webhook delivery is logged automatically. In the webhook detail view, you can see:

• All recent deliveries with their timestamp and status (success or failure)
• Which attempt number each delivery was (1 through 5)
• Response times

If a webhook's success rate drops significantly, the system alerts the administrator automatically.

Automatic retry behavior

When a delivery fails (the external system is temporarily unavailable or returns an error), the system retries automatically:

After 5 failed attempts, the webhook is flagged and the administrator is notified. The webhook is not automatically disabled — deliveries continue for new events.

AI-assisted configuration

If you are not sure which events to subscribe to or how to configure the webhook, describe what you want in plain language and the system will suggest the configuration:

1. Click Configure with AI.
2. Describe the use case (e.g., "Notify our Slack #alerts channel when a critical maintenance task is created").
3. Review the suggested configuration — events, filters, and payload format.
4. Accept or adjust and save.

Key benefits

• Real-time push notifications to any external system — no polling required
• Automatic retries ensure delivery even if the external system is temporarily unavailable
• Delivery log for troubleshooting and audit of what was sent and when
• Test mode to verify configuration without generating real events
• AI-assisted setup for quick configuration without technical expertise
• Filters to forward only the events that matter for each integration

Common use cases

Scenario 1: Critical alarms to Slack The operations team wants to know immediately when a Critical or Emergency alarm fires, even if they are not in the Rela AI dashboard. A webhook is configured pointing to the Slack #critical-alerts channel webhook URL, filtered to only Critical and Emergency severity events. When compressor C-02 generates a critical temperature alarm, the Slack channel receives a message within seconds: "CRITICAL ALARM — Compressor C-02: temperature 92°C. Agent created task MAN-0456."

Scenario 2: Sync completed tasks to external CMMS The plant uses IBM Maximo as its CMMS for compliance reporting. When a maintenance task is completed in Rela AI, the Maximo record must be updated. A webhook fires when a task status changes to "done," sending the task details to a Maximo API endpoint. The CMMS stays synchronized without manual data entry.

Scenario 3: Incident management integration The IT team manages infrastructure incidents in PagerDuty. A webhook is configured to forward any Emergency severity event from the "Utilities" source group to a PagerDuty integration endpoint. When a power supply failure event arrives at 3 AM, PagerDuty creates an incident and pages the on-call engineer — without anyone having to monitor Rela AI directly.