Vercel Security Review Now Works Without Linked Projects, Covering All GitHub Repositories

Vercel Security Review now scans code changes in GitHub repositories not linked to projects.
Vercel announced an update to Security Review: vulnerabilities in code changes can now be scanned without linking a GitHub repository to a Vercel project, greatly lowering the barrier to use. Previously tied to projects and deployments, reviews are now triggered automatically on push for any repository accessible via the Vercel GitHub App. Coverage defaults to all accessible unlinked repositories and can be narrowed by account or repository in Agent settings, while linked projects still benefit from deeper deployment-context analysis.
Vercel's Security Review feature has received a major update: you can now scan code changes for vulnerabilities without linking a GitHub repository to a Vercel project. This change significantly lowers the barrier to using Security Review, allowing more repositories to benefit from automated code security scanning.

From "Project Binding" to "Direct Repository Scanning"
Previously, Vercel's Security Review was typically tied to linked projects and deployment workflows. This meant that if a repository wasn't yet linked to a Vercel project, it couldn't take advantage of this security capability. For many repositories in early stages, or used only for internal tools and experimental code, this created an invisible barrier.
According to Vercel's official documentation, Security Review can now inspect commits directly in repositories where the feature is enabled, without relying on a Vercel project or deployment. When a push creates a deployment for a linked project with Security Review enabled, the review also leverages additional project and deployment context to provide richer analysis. This design preserves the advantages of deep integration while giving unlinked repositories the same level of security protection.
Default Coverage and Fine-Grained Control
In terms of coverage, Vercel adopts a "broad by default, narrow on demand" strategy. By default, Security Review covers all unlinked repositories that the Vercel GitHub App can access in connected GitHub accounts. For organizations that want precise control over review scope, coverage can be limited in Agent settings to selected connected accounts or individual repositories.
This flexibility is particularly important for teams with large numbers of repositories. Teams can prioritize enabling reviews for critical business repositories and avoid unnecessary scanning overhead for temporary repositories. Once enabled, every push to a covered repository automatically triggers a review of the corresponding code changes, with no additional configuration or manual action required.
Enablement Process and Integration
The enablement process is straightforward. Users first need to connect a GitHub account in the team's Git settings, then enable Security Review in Vercel Agent settings. The feature directly uses the GitHub App's access permissions, so connecting an account is a prerequisite.
Interestingly, this update transforms Security Review from "an appendage of the deployment pipeline" into "a standalone capability of code repositories." Developers no longer need to set up a complete Vercel project just to get security scanning. This decoupling significantly improves practicality for teams that develop across multiple platforms or use Vercel as a supplementary tool rather than their primary hosting platform.
Vercel GitHub App is the official integration between Vercel and GitHub, installed to a GitHub account or organization via OAuth authorization. The permissions it requests typically include reading repository contents, commit statuses, pull requests, and more, to support capabilities such as deployments, preview environments, and security reviews. Security Review directly reuses the App's access token to read code changes, so "connecting a GitHub account in Git settings" essentially completes this authorization chain; without authorization, Vercel cannot detect repository pushes and therefore cannot trigger reviews.
Security Review and Vercel Agent: Vercel Agent is an intelligent agent layer within the Vercel platform for automated tasks (such as code review and security analysis). As one of its capabilities, Security Review's toggle and coverage scope are centrally managed in Agent settings rather than scattered across individual project deployment configurations. This centralized control allows organizations to unify policies across repositories, and also explains why repositories without linked projects can still be scanned—the scheduling entity for reviews has moved up from "projects" to the "Agent."
Impact on Development Workflows
From a workflow perspective, this change brings security review closer to the code itself. The philosophy of shift-left security emphasizes finding vulnerabilities early, and binding reviews directly to push events is a concrete implementation of this principle. After developers submit code, reviews run automatically, allowing feedback to enter the development loop earlier.
For teams already deploying with Vercel, Security Review can still provide deeper analysis combined with project context. For repositories not yet linked to projects, basic change-level scanning is now available. This tiered design balances both depth and breadth.
Overall, this Vercel update lowers the barrier to enabling Security Review, expands coverage, and through configurable scanning policies lets teams find a balance between security and efficiency. For GitHub users who value code security, this is an improvement worth paying attention to.
Shift Left Security refers to introducing security checks as early as possible in the software development lifecycle, rather than waiting until before release or after going live. Binding scanning to push events is a typical shift-left practice: vulnerabilities are discovered as soon as code enters the repository, and the cost of fixing them is far lower than later stages. A related concept is diff-based scanning, which analyzes only the code differences involved in the current commit rather than the entire codebase. This shortens feedback time and reduces false-positive noise; the trade-off is that it may miss issues in existing code, so it is often used alongside periodic full scans.
One-line Summary
Vercel Security Review now supports directly scanning code changes in GitHub repositories not linked to projects.
Paragraph Summary
Vercel announced an update to its Security Review feature: vulnerabilities in code changes can now be scanned without linking a GitHub repository to a Vercel project, significantly lowering the barrier to use. Previously, security reviews were tied to projects and deployment workflows, so unlinked repositories couldn't use them; now, as long as a repository is accessible through the Vercel GitHub App, pushing code automatically triggers a review. By default, all accessible unlinked repositories are covered, and scope can be narrowed by account or individual repository in Agent settings. Linked projects can still provide deeper analysis combined with deployment context. To enable it, first connect a GitHub account in the team's Git settings, then turn on the feature in Vercel Agent settings. This decoupling brings security review closer to the code itself and reflects the shift-left security philosophy.
Related articles

Zed Editor Releases v1.29.2: Observations on a Routine Version Iteration
Zed editor releases patch v1.29.2, version bump by @JosephTLyons, commit verified via GitHub signature. This article covers the release's basic info, what a patch version means, and practical upgrade advice for users.

Anthropic IPO Questioned: Hacker News Debates Timing of Going Public
A Hacker News post titled "Just say no to Anthropic's IPO" sparks debate on the timing of Anthropic's public listing and AI capitalization, touching on valuation bubbles and mission independence.

Hands-On with the Genie 4 World Model: Directing Worlds in Real Time Like a Puppeteer
Google DeepMind's Genie 4 world model in depth: Special Guest X's hands-on demo of real-time scene direction via prompts, custom interactive apps in Project Genie, and prompts-as-seeds reproducibility.