Skip to main content

​
Overview

Rootly AI is an AI responder built into the places your team already works. Ask it what’s happening and it answers from the incident’s live context. In user-initiated conversations, Rootly-native reads and actions are capped at what you can access in Rootly, and native actions are audited under the name of the person who asked. Connector reads and actions follow that connector’s Rootly permission checks and the access of its connected provider account. Automatic AI SRE runs instead use the background execution boundary described in Investigation Rules. Atlas is the AI layer behind these experiences. AI SRE investigations, Slack mentions, web and mobile chat, and the Chat API are different entry points with different tools and access rules. The Chat API is enabled separately per team and also requires Rootly Agent in Slack to be on; see Chat API endpoints and availability. Most Rootly AI features are opt-in and governed from AI Settings. Rootly AI SRE is enabled for the current team by your Rootly account team, then managed under AI SRE → Atlas, or under AI & Agents where that navigation is still in use. See Where AI Settings Live.

​
Where It Works

​
What It Does

​
Extend It

​
Govern It

AI Settings governs the self-service Rootly AI features and surfaces the current team can use. Your Rootly account team enables AI SRE for that team; after enablement, its instructions, investigation rules, and memory are managed under AI SRE → Atlas (AI & Agents → AI SRE if your sidebar doesn’t have an AI SRE item).