EveStudioDocs

Deploy to Vercel

Deploying ships your agent to production, where it runs as Vercel Functions and its schedules actually fire. Studio's Deploy tab drives the whole thing — build, deploy, environment, and the logs — from one place.

Before you deploy

Deploying requires the agent to be connected to Vercel. That's the same link that lets the model run: it associates the agent's folder with a Vercel project and pulls the AI Gateway credential. If you haven't done it yet, do it first — see Connect to Vercel. Without a linked project there's nothing to deploy to.

Build and Deploy

Open an agent and select the Deploy top tab. Two buttons drive the pipeline:

  • Build — compiles the agent locally so you can catch errors before shipping.
  • Deploy — runs eve deploy, which builds and pushes the agent to your linked Vercel project as a production deployment.

Both stream their output into a command-output console in the tab. Deploys take a while — installing, building, and uploading — so the console shows a live "working…" indicator with an elapsed timer while a step runs, and prints the result (including the deployment URL) when it finishes. Alongside it, a dev-server logs pane shows the local eve dev server's output, which is where you look when the local agent misbehaves.

Environment and Sandbox

Near the deploy controls, the Environment panel manages the agent's configuration:

  • Edit .env / .env.local — the local environment files the dev server reads.
  • Pull / push Vercel env vars — sync environment variables between your machine and the linked Vercel project. Pull brings the project's variables down (this is how the AI Gateway token lands locally); push sends local values up so the deployed agent has them.
  • Secrets — keys like API tokens live here as environment variables, never in code. (For example, an Arcana memory key is stored as an env var and pushed to the project.)

A Sandbox section sits alongside — the isolated runtime the agent uses for file and command work.

Chat against the deployed agent

Once deployed, you don't need to leave Studio to try it. In the Chat tab, the Target toggle switches between:

  • Local — the eve dev server on your machine.
  • Deployed — the production agent you just shipped.

Flip Target to Deployed and your messages hit the live Vercel deployment, streaming the same turns, tool calls, and approvals as local chat.

Protected deployments

If your Vercel project has Deployment Protection enabled, the production URL is gated and Studio can't reach it by default. Generate a Protection Bypass token in your Vercel project settings and paste it into Studio, so the app can authenticate to the protected deployment and chat against it.

Where to go next