.svg)
Rootly Agent is now built into the web app. Open the chat panel from any incident and ask questions in plain language, and it answers using that incident's live context: timeline, severity, status, roles, alerts, and linked data. You get grounded answers without piecing them together yourself.
Use it to get caught up, see who is doing what, understand why an incident is a given severity, review what has been tried, or draft communications for different audiences. If Rootly Meeting Scribe captured the bridge call, you can ask for a recap of what was discussed. Replies stream in as they are generated and are formatted in markdown, with a copy button so you can drop a drafted update straight into Slack, email, or a doc.
What's included
- A chat panel on every incident page, docked alongside the incident so the rest of the page stays usable
- Suggested prompts to get started, plus natural follow-up questions that keep their context
- Answers grounded in live incident data: timeline, severity, status, roles, alerts, and linked records
- Drafting help for status updates, executive summaries, and customer messages
- Multiple chats per incident with automatic naming and history, private to you
- Times shown in your team's time zone by default, with on-demand conversion to UTC or another zone
Why it matters
During an incident the context you need is scattered across the timeline, the alerts, and the bridge call. Rootly Agent brings it into one place and answers in plain language, so responders spend less time hunting for status and more time resolving the incident. On the web it is read-only today; to page, change status or severity, or manage action items, use @Rootly in Slack or the incident page directly.
Getting started
Rootly Agent in Web is available to all customers and can be turned on by an Admin under Configuration → Rootly AI. Using the panel requires an Incident Response seat. See the documentation to learn more.
