Vercel Agent is now available in Slack code channels
Vercel Agent can now investigate incidents, edit code, and review PRs inside shared Slack code channels, with approval gates and an audit trail for every action.
Vercel Agent now runs in **Slack code channels**, carrying context from an existing channel or thread and posting deployments, logs, errors, diffs, and PR findings as it works.
Teams can use one shared conversation for incidents, migrations, and reviews. The agent is **read-only by default**, drafts a plan before changes, and records each request, approval, and action.
Vercel Agent now runs in **Slack code channels**, carrying context from an existing channel or thread and posting deployments, logs, errors, diffs, and PR findings as it works. Teams can use one shared conversation for incidents, migrations, and reviews. The agent is **read-only by default**, drafts a plan before changes, and records each request, approval, and action. The integration is a **public beta for Pro and Enterprise teams**. The material does not describe beta limits or how reliably the agent distinguishes instructions from surrounding conversation.
This turns shared Slack discussion into both context and control surface for coding work, incidents, migrations, and reviews. Against prior channel-based agent concepts, it adds concrete governance signals—read-only default, plans, approvals, and action records—but remains a public beta with unresolved limits and instruction-boundary reliability, so shared context should not be treated as automatically safe authority.