Vercel Connect now supports Custom Environments, letting you link a connector to any named environment beyond the default Production, Preview, and Development tiers. Pro and Enterprise plans get access. Connect itself remains in beta across all plans.
The mechanics are straightforward but the details matter. From the CLI, pass the environment slug via --environment when deploying. Deployments in that environment call getToken as normal. If the calling environment is not on the project link, the request fails with ClientNotEnabledForEnvironmentError. Trigger destinations can also target a Custom Environment, but you must deploy to it and assign a verified domain first. Vercel Connect forwards webhook events through that domain.
The security model has a hard limit worth understanding: project links control which deployments can request tokens, but they do not restrict a token after it is issued. If you need provider-level isolation between environments, Vercel's own guidance says to create a separate connector per environment and scope it to only the permissions it needs. That distinction between request-time control and post-issuance control is the real reason to read the full documentation.
[READ ORIGINAL →]