# How Vercel Agent accesses production safely

DevFeed: [How Vercel Agent accesses production safely](<https://devfeed.tech/articles/how-vercel-agent-accesses-production-safely-97699.md>)

Original publisher: [Read original article](<https://vercel.com/blog/how-vercel-agent-accesses-production-safely>)

Author: Tom Dale

Published: 2026-10-23T17:23:19.538589Z

Content type: article

Language: en

Sources: [Vercel News](<https://devfeed.tech/sources/vercel-news.md>)

Topics: [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [human-in-the-loop UX](<https://devfeed.tech/topics/human-in-the-loop-ux.md>), [Authorization](<https://devfeed.tech/topics/authorization.md>), [least privilege](<https://devfeed.tech/topics/least-privilege.md>), [Vercel](<https://devfeed.tech/topics/vercel.md>), [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [authorization](<https://devfeed.tech/tags/authorization.md>), [automatically](<https://devfeed.tech/tags/automatically.md>), [build](<https://devfeed.tech/tags/build.md>), [capabilities](<https://devfeed.tech/tags/capabilities.md>), [compare](<https://devfeed.tech/tags/compare.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [deployments](<https://devfeed.tech/tags/deployments.md>), [failed](<https://devfeed.tech/tags/failed.md>), [inspect](<https://devfeed.tech/tags/inspect.md>), [investigation](<https://devfeed.tech/tags/investigation.md>), [logs](<https://devfeed.tech/tags/logs.md>), [permission](<https://devfeed.tech/tags/permission.md>), [permissions](<https://devfeed.tech/tags/permissions.md>), [plan](<https://devfeed.tech/tags/plan.md>), [production](<https://devfeed.tech/tags/production.md>), [production-deployment](<https://devfeed.tech/tags/production-deployment.md>), [scope](<https://devfeed.tech/tags/scope.md>), [scopes](<https://devfeed.tech/tags/scopes.md>), [vercel](<https://devfeed.tech/tags/vercel.md>)

## AI overview

Vercel Agent investigates production issues under a read-only baseline, then requests approval for a plan when work needs broader access or a supported external action. Plans link the user's intent to proposed tasks and requested scopes; for example, reading deployment logs can proceed automatically, while redeploying a build requires deployment write access and plan approval. Approved credentials remain limited by scope, team, the user's permissions, and expiry, with Vercel APIs checking each request. GitHub pull requests use separate credentials tied to a specific repository. The runtime verifies that a plan is active before using its credential, including when resuming work; stopped plans cannot use it. Shared scope definitions let planning reject unknown or outdated permissions, though the cached catalog can lag registry updates. The approach still has limits: scopes can grant more authority than a particular action needs, read-only work can expose sensitive information, and permitted actions can cause harm. The text says scoped credentials work alongside sandboxing, network policy, secret handling, rate limits, audit logs, and service authorization.

## Source excerpt

An agent working on production systems needs permission to act. A request like "fix this deployment" doesn't tell you upfront which actions it will take or what access it will need. Grant agents broad access and a mistake can affect unrelated resources. Ask for approval at every step and the user has to approve individual actions before they know how the agent intends to solve the problem. Repeated prompts can turn that review into a habit of clicking approve. Vercel Agent starts by investigating the problem, then proposes a plan showing what it would do and what access it would need. Approving that plan grants the additional permissions to carry it out, so the user can review the proposed fix before authorizing any changes. The plan determines the permissions For a broken production deployment, Vercel Agent might inspect recent deployments and build logs, then compare the failed deployment with the last known-good one. Read-only access lets that investigation run automatically. If it leads to a proposed redeploy, Vercel Agent requests the additional access in a plan the user can review and approve in one decision. Access is defined by scopes, the named capabilities recognized by the Vercel authorization system. Any plan requesting access beyond the read-only baseline waits for approval as a whole, including any read-only tasks it contains. In this example, the investigation runs under a separate read-only plan before Vercel Agent proposes the redeploy: Plan Work Requested scope Investigate the failure Inspect recent deployments and build logs, then compare the failed deployment with the last successful one read:deployment Redeploy the last known-good build Create a deployment read-write:deployment A short read-only lookup can run without a plan. For longer investigations, Vercel Agent creates a read-only plan so the user can follow its tasks and progress while it runs automatically. The plan connects what the user wants done to what Vercel Agent is allowed to do: I