Sovra Recon·Glossary · Open redirect

Open Redirect: Definition, Risks, and Protection

By Solan Barlet, co-founder of Sovra Recon
Updated October 29, 2026

In short

An open redirect occurs when an application redirects the user to a URL provided as a parameter without validating it. An attacker can then create a link that starts with your domain—and is therefore trusted—but directs the victim to a malicious site. Often considered a minor vulnerability on its own, it is used for phishing and, when combined with other vulnerabilities, can enable the theft of OAuth authentication tokens.

An example

After logging in, an application redirects the user to the requested page with a URL such as /login?next=/account. If it also accepts /login?next=https://site-piege.example, a link sent via email will display the legitimate domain, but the victim will end up on a copy of the site that asks them to re-enter their password.

Why This Is a Real Risk

  • Phishing: The link appears trustworthy and bypasses certain filters because it points to a known domain.
  • OAuth token theft: If the identity provider accepts a callback URL on your domain, an open redirect can leak the code or token to the attacker.
  • Filter bypass: A domain whitelist (such as anti-SSRF protection) can be bypassed by routing through a redirect hosted on an authorized domain.

How to Protect Against It

  • Avoid redirects to a user-supplied URL; instead, use an identifier associated with a server-side defined destination.
  • Otherwise, accept only internal relative paths or domains on a strict whitelist.
  • Be wary of common workarounds: addresses starting with //, misleading domains (your-domain.fr.trap-site.example), and URL encodings.
  • Display a warning page before leaving the site when an external redirect is necessary.

Frequently asked questions

Is an open redirect a serious vulnerability?

On its own, it is generally classified as low to medium severity. It becomes serious when combined with an OAuth flow, an anti-SSRF filter, or a targeted phishing campaign.

Sources

Find out where an attacker would get in.

Our AI agents actually attack your scope, fully automated, and hand you the proof of every vulnerability.