Vercel Agent Can Now Validate Alerts Before Paging Your Team

How alert validation works

Validation is configured per webhook destination. According to Vercel, users open a monitor's notification settings and create or link a webhook destination for their paging system, such as incident.io. They then select Critical Only next to that destination.

Once that is set, Agent examines each alert before it is sent on. Vercel says Agent checks logs and metrics to assess impact, and only alerts it rates as Critical are passed to the paging destination.

What Agent looks for

Vercel gives two examples of the kind of judgment Agent applies. It can identify health-check errors that do not affect customers, which would then not be treated as Critical. It can also confirm sustained 5xx errors affecting checkout requests, a case that would warrant a page.

Users can also add context to each monitor to guide Agent's investigation. This lets teams describe what matters for a given service, so the assessment reflects their own setup rather than relying only on the raw signals.

Availability and pricing

Alert validation is available on Pro and Enterprise plans with Observability Plus. The details from Vercel:

• Observability Plus includes 10 Agent investigations per billing cycle.
• Additional investigations are billed separately.
• Custom monitors cost $0.0035 per GB of data scanned.
• Built-in alerts remain included with Observability Plus.

Vercel points to its alert webhook documentation for further setup details.

What it means for teams

The feature targets a common on-call problem: pages triggered by issues that do not affect customers. By filtering alerts through Agent before they reach a paging tool, teams can limit pages to incidents that Agent judges to be critical. Because each investigation counts toward the plan's allowance of 10 per billing cycle, with extras billed separately, teams will want to consider how many alerts they expect to validate.