AI and agents are creating new ways in and making vulnerabilities easier to exploit. Shipping faster shouldn’t mean taking on more risk
Vulnerability detectedMedium12mPrototype pollution in config parser
Vulnerability detectedHigh4mServer-side request forgery in webhook proxy
Vulnerability detectedCriticalVercel Agent just fixed thisRemote code execution in file upload handler
Build a customer portal with sign-in and payments|
A prompt asks to build a customer portal with sign-in and payments. Above it, vulnerability alerts keep stacking up: a critical remote code execution, a high-severity server-side request forgery, and a medium prototype pollution issue. Vercel Agent has already fixed the critical one.
Reduce risk, right in your workflow
From automatically blocking compromised dependencies to scanning every pull request for vulnerabilities, Vercel makes shipping secure code faster
SQL injection:id is interpolated into the query. Pass it as a parameter so it can’t run as SQL.
-const user = await db.query(
- `SELECT * FROM users WHERE id = ${id}`
-)
+const user = await db.query(
+ 'SELECT * FROM users WHERE id = $1',
+ [id],
+)
Vercel flags a SQL injection vulnerability on a pull request and replaces the query built from user input with a parameterized query.
Secure execution for every agent
Give agents the autonomy to execute code and interact with external systems, backed by Vercel’s integrated infrastructure for isolated execution, secure connectivity, and scoped access
A user asks data-agent to pull a customer export from Snowflake, install Python packages, run a churn analysis, and build a dashboard. The request passes through Sandbox, Connect, OIDC Federation, and Secure Compute before the agent replies with a churn analysis notebook and a link to the live dashboard.
Runtime defenses which ship with every app and agent
Defend against malicious traffic, DDoS attacks, and unwanted bots with security tailored to how your applications and agents behave
Three bot requests and one user request reach Vercel Firewall. Each bot request is blocked individually at the firewall, while the user request passes through to the protected app.
Your entire security posture, in one view
See where you’re exposed and what needs attention, with security checks across your accounts, projects, and deployments brought together in one place
Assign roles to give team members the permissions they need to work on your projects.
Deployment Protection
Secure your Vercel project's preview and production URLs with fine-grained access control.
Audit Logs
Enterprise
Track and analyze your team members' activity. Accessed by team members with the owner role.
Directory Sync
Enterprise
Manage your organization's memberships by integrating with third-party identity providers.
ISO 27001
SOC 2
PCI DSS
HIPAA
GDPR
DPF
Frequently asked questions
01
Is Vercel TISAX Compliant?
Yes, Vercel has completed TISAX Assessment Level 2 to serve the automotive industry. Customers can access results directly through the ENX portal. To view the assessment details, sign in to your account on the ENX portal and search for Vercel, or look up the following details: Assessment ID: AMR06H-1; Scope ID: SYN3TM.
02
Is Vercel certified under the Data Privacy Framework (DPF)?
Yes. Every plan supports Deployment Protection, including Vercel Authentication, Shareable Links (Hobby is limited to one per account), and deployment protection exceptions. Pro and Enterprise teams can also require a password with Password Protection.
04
Does Vercel encrypt data?
Yes. Data is encrypted at rest (AES-256) and in transit (HTTPS / TLS), including sensitive information like access tokens and secrets.
05
Does Vercel backup the data on its platform?
Yes. Our current backup interval is every two hours and each backup is persisted for 30 days. Automatic backups are taken without affecting the performance or availability of the database operations. All the backups are stored separately in a storage service, and those backups are globally replicated for resiliency against regional disasters. If a database instance is deleted, all associated backups are also automatically deleted. Backups are periodically tested by the Vercel engineering team.
06
What infrastructure does Vercel use?
The Vercel CDN & deployment platform primarily uses Amazon Web Services (AWS). In the case of an AWS outage, our network is resilient to regional downtime. Vercel will automatically route traffic to the nearest available edge. Vercel.com uses Azure CosmosDB to store and globally replicate data, which is different than our CDN. This is an additional step taken to ensure uptime for applications on our platform.
07
Does Vercel provide infrastructure segregation?
Enterprise Teams on Vercel have their own build infrastructure ensuring isolation from Hobby/Pro accounts on Vercel.
08
Does Vercel conduct regular penetration testing and vulnerability scans?
Yes. Vercel conducts regular penetration testing with third-party experts. In addition to our annual penetration tests, we consistently perform targeted assessments on an ongoing basis. We also implement daily code reviews, static analysis checks, and dependency scanning at the code level. Our cloud security posture management platform (CSPM) facilitates workload vulnerability scanning. Pro and Enterprise customers can request access to our latest annual penetration testing reports.
09
Does Vercel use subprocessors?
Yes, a list of our current subprocessors can be found on our subprocessors page.
10
Does Vercel have a bug bounty program?
Yes. Vercel has a public HackerOne bug bounty program that rewards researchers for finding and reporting security vulnerabilities. To report a vulnerability, please submit through our HackerOne page.
11
Does Vercel protect against OWASP Top 10?
Yes. Vercel offers managed rulesets, including one specifically designed to protect against the OWASP Top 10 risks. This feature is available on Enterprise plans.