Anomaly alerts just got a terminal-native workflow. Vercel has added CLI access to alert details, meaning you no longer need to open a dashboard to see what is firing on your project or team.
The new vercel alerts command lists all alerts scoped to a team or a specific project. For each alert, you get the start time, the alert type, and whether the alert is still active. That is enough context to triage quickly without context-switching to a browser.
The more interesting addition is the AI investigation flag. When you pass that option, AI investigation results appear alongside each alert in the output. That means the diagnostic reasoning lives right next to the raw alert data in your terminal. You can read it, pipe it, or hand it to an agent in your toolchain.
That last point matters more than it might look. Agents running automated incident workflows previously had no clean programmatic path to Vercel anomaly context. Now they do. A coding agent or ops automation script can call vercel alerts, read the AI investigation output, and take action, all without a human opening a browser tab.
There are no new services to configure and no API keys to wire up separately. If you already have the Vercel CLI installed and authenticated, the command is available now.
What should you do with this today? If you run on-call rotations or have any automated incident response scripts, update them to call vercel alerts with the AI investigation option. Pull the output into whatever alerting or chat pipeline your team already uses. If you are building an agent that monitors production deployments, this is a concrete new tool input worth adding to its context window.