CVE-2026-33495
MEDIUMOry Oathkeeper has an authentication bypass by usage of untrusted header
EPSS Exploitation Probability
EPSS (Exploit Prediction Scoring System) is a daily probability model maintained by FIRST.org. It estimates the likelihood a CVE will be exploited in production environments within the next 30 days, derived from real-world threat intelligence signals.
Blast Radius
github.com/ory/oathkeeperReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects Go packages — download data is not available via public APIs for these ecosystems.
Description
ORY Oathkeeper is an Identity & Access Proxy (IAP) and Access Control Decision API that authorizes HTTP requests based on sets of Access Rules. Ory Oathkeeper is often deployed behind other components like CDNs, WAFs, or reverse proxies. Depending on the setup, another component might forward the request to the Oathkeeper proxy with a different protocol (http vs. https) than the original request. In order to properly match the request against the configured rules, Oathkeeper considers the X-Forwarded-Proto header when evaluating rules. The configuration option serve.proxy.trust_forwarded_headers (defaults to false) governs whether this and other X-Forwarded-* headers should be trusted. Prior to version 26.2.0, Oathkeeper did not properly respect this configuration, and would always consider the X-Forwarded-Proto header. In order for an attacker to abuse this, an installation of Ory Oathkeeper needs to have distinct rules for HTTP and HTTPS requests. Also, the attacker needs to be able to trigger one but not the other rule. In this scenario, the attacker can send the same request but with the X-Forwarded-Proto header in order to trigger the other rule. We do not expect many configurations to meet these preconditions. Version 26.2.0 contains a patch. Ory Oathkeeper will correctly respect the serve.proxy.trust_forwarded_headers configuration going forward, thereby eliminating the attack scenario. We recommend upgrading to a fixed version even if the preconditions are not met. As an additional mitigation, it is generally recommended to drop any unexpected headers as early as possible when a request is handled, e.g. in the WAF.
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐹Go | github.com/ory/oathkeeper | all versions | 0.40.10-0.20260320084810-e9acca14a04d |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for github.com/ory/oathkeeper. O3's reachability analysis confirms whether the vulnerable code path is actually invoked in your application, so you act on real exposure instead of every transitive match.
Fix
Update github.com/ory/oathkeeper to 0.40.10-0.20260320084810-e9acca14a04d or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms CVE-2026-33495 is resolved across your whole dependency graph.
Workarounds
If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.
How O3 protects you
O3 pinpoints whether CVE-2026-33495 is reachable in your code and exactly where to fix it, then blocks exploitation in production at runtime until the patched version is deployed.
Tailored to CVE-2026-33495. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.
Frequently Asked Questions
Is CVE-2026-33495 in your dependencies?
O3 detects CVE-2026-33495 across Go dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.