Browse / AI Framework and Platform Flaws / CVE-2025-49592
n8n allows open redirects via the /signin endpoint
- Class
- AI Framework and Platform Flaws (filed by a model that read the advisory)
- Technique
- Other weaknesses
- Affected
- n8n (npm)
- CVSS
- CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:U/C:L/I:L/A:N
- EPSS
- 0.2% chance of exploitation in 30 days
How it works, in short
The post-login redirect parameter on the sign-in page is not restricted to same-origin paths, so users can be sent to attacker domains. (Our summary, written by a model. The source text is below.)
Details
Impact
This is an Open Redirect (CWE-601) vulnerability in the login flow of n8n. Authenticated users can be redirected to untrusted, attacker-controlled domains after logging in, by crafting malicious URLs with a misleading redirect query parameter.
This may lead to:
- Phishing attacks by impersonating the n8n UI on lookalike domains (e.g., n8n.local.evil.com)
- Credential or 2FA theft if users are tricked into re-entering sensitive information
- Reputation risk due to the visual similarity between attacker-controlled domains and trusted ones
The vulnerability affects anyone hosting n8n and exposing the /signin endpoint to users.
Patches
The issue has been patched in 1.98.0.
All users should upgrade to this version or later.
The fix introduces strict origin validation for redirect URLs, ensuring only same-origin or relative paths are allowed after login.
Patch commit: https://github.com/n8n-io/n8n/pull/16034
References
- https://github.com/n8n-io/n8n/security/advisories/GHSA-5vj6-wjr7-5v9f
- https://nvd.nist.gov/vuln/detail/CVE-2025-49592
- https://github.com/n8n-io/n8n/pull/16034
- https://github.com/n8n-io/n8n/commit/4865d1e360a0fe7b045e295b5e1a29daad12314e
- https://github.com/n8n-io/n8n
- https://github.com/n8n-io/n8n/releases/tag/n8n%401.98.0
Source: GitHub Advisory Database, used under CC BY 4.0.