---
id: "GHSA-prmc-fg87-r4gh"
page: "https://infrastack.ai/live/GHSA-prmc-fg87-r4gh"
kind: "vulnerability"
aliases: ["GHSA-prmc-fg87-r4gh"]
title: "Credential-Using Nodes Forward Authentication Headers on Cross-Origin Redirects"
severity: "medium"
cvss: 6.3
epss: null
exploited: false
status: "open"
category: "ai-tooling"
products: [{"name": "n8n", "fixed": [">= 2.46.1"], "affected": "< 2.46.1", "ecosystem": "npm"}, {"name": "n8n", "fixed": [">= 2.45.5"], "affected": "< 2.45.5", "ecosystem": "npm"}, {"name": "n8n", "fixed": [">= 3.0.1"], "affected": "< 3.0.1", "ecosystem": "npm"}]
ecosystems: ["npm"]
tags: ["ai-tooling", "http-request", "n8n", "self-hosted", "workflows"]
not_affected_if: ["default-config-no-untrusted-urls", "trusted-users-only"]
published_at: "2026-10-28T08:49:03+00:00"
first_seen_at: "2026-11-09T22:53:25+00:00"
updated_at: "2026-11-09T22:55:01+00:00"
sources: [{"source": "ghsa-repo", "url": "https://github.com/n8n-io/n8n/security/advisories/GHSA-prmc-fg87-r4gh", "first_seen_at": "2026-11-09T22:53:25+00:00"}]
exploit_sources: []
---

# Credential-Using Nodes Forward Authentication Headers on Cross-Origin Redirects

Link for people: https://infrastack.ai/live/GHSA-prmc-fg87-r4gh (this .md is for agents)

n8n re-sent the Authorization header and basic auth credentials to a different origin when an authenticated HTTP request was redirected there. An attacker who can steer a redirect target, through an expression-fed URL or a malicious endpoint, can collect the stored credential.

_Summary, check and fix written by a model from the sources; labelled as such._

## Affected

- npm:n8n affected < 2.46.1 fixed >= 2.46.1
- npm:n8n affected < 2.45.5 fixed >= 2.45.5
- npm:n8n affected < 3.0.1 fixed >= 3.0.1

## How to check

- Check the installed n8n version with `n8n --version` or the `n8n` entry in `package.json`; versions below `2.45.5`, `2.46.1` or `3.0.1` are affected depending on the release line
- Look for workflows that use credential-authenticated nodes (other than HTTP Request v4.4) with URLs built from untrusted input

## Fix

- n8n `2.45.5` on the 2.45 line
- n8n `2.46.1` on the 2.46 line
- n8n `3.0.1` on the 3.x line

## Timeline

- 2026-11-09T22:53:25+00:00 ghsa-repo https://github.com/n8n-io/n8n/security/advisories/GHSA-prmc-fg87-r4gh

## Documents

- [advisory] Maintainer's advisory: github.com https://github.com/n8n-io/n8n/security/advisories/GHSA-prmc-fg87-r4gh

## From the sources

## Impact

When an HTTP request made by a credential-authenticated node followed a redirect to a different origin, n8n re-attached the `Authorization` header and basic auth credentials to the redirected request. The internal helper that builds the redirect configuration defaulted `sendCredentialsOnCrossOriginRedirect` to `true`. The HTTP Request node v4.4 opted out of this default, but other credential-using nodes did not. Where a workflow makes authenticated HTTP requests and the redirect target can be influenced by untrusted data, such as an expression-fed URL or a redirect issued by an attacker-controlled endpoint, the stored credential could be forwarded to an unintended destination.

## Patches

The issue has been fixed in n8n versions 2.46.1, 2.45.5 and v3.0.1. Users should upgrade to one of these versions or later to remediate the vulnerability.

## Workarounds

If upgrading is not immediately possible, administrators should consider the following temporary mitigations:
- Restrict n8n instance access to fully trusted users only.
- Audit active workflows that use credential-authenticated HTTP requests and ensure that request URLs and any data that could influence redirect targets do not incorporate untrusted external input.
- Apply network-level controls to restrict outbound HTTP requests from the n8n process to known, trusted destinations.
- Enable SSRF protection by setting `N8N_SSRF_PROTECTION_ENABLED=true` to reduce the risk of attacker-influenced outbound requests.
- Rotate any credentials used by workflows that may have been exposed to untrusted redirect targets.

These workarounds do not fully remediate the risk and should only be used as short-term mitigation measures.

---

_Free to read, with fair use of about 60 pages a minute. For higher limits or API access: sales@infrastack.ai_