OAuth2-Proxy vulnerable to header smuggling via underscore, leading to potential privilege escalation
Not confirmed exploited. 1 of 8 rule sources name it.
OAuth2-Proxy is an open-source tool that can act as either a standalone reverse proxy or a middleware component integrated into existing reverse proxy or load balancer setups. In versions prior to 7.13.0, all deployments of OAuth2 Proxy in… More
OAuth2-Proxy is an open-source tool that can act as either a standalone reverse proxy or a middleware component integrated into existing reverse proxy or load balancer setups. In versions prior to 7.13.0, all deployments of OAuth2 Proxy in front of applications that normalize underscores to dashes in HTTP headers (e.g., WSGI-based frameworks such as Django, Flask, FastAPI, and PHP applications). Authenticated users can inject underscore variants of X-Forwarded-* headers that bypass the proxy’s filtering logic, potentially escalating privileges in the upstream app. OAuth2 Proxy authentication/authorization itself is not compromised. The problem has been patched with v7.13.0. By default all specified headers will now be normalized, meaning that both capitalization and the use of underscores (_) versus dashes (-) will be ignored when matching headers to be stripped. For example, both `X-Forwarded-For` and `X_Forwarded-for` will now be treated as equivalent and stripped away. For those who have a rational that requires keeping a similar looking header and not stripping it, the maintainers introduced a new configuration field for Headers managed through the AlphaConfig called `InsecureSkipHeaderNormalization`. As a workaround, ensure filtering and processing logic in upstream services don't treat underscores and hyphens in Headers the same way.
Edge coverage
Claimed by vendorsAsk Vigil
AI · answers from this pageWorkers AI (Llama 3.3). Can be wrong: check the cited source.
Severity, decoded
CVSS 3.1Exploitable over the network, with low privileges and no user interaction, leading to full loss of confidentiality and low impact on integrity, and the impact can reach components beyond the vulnerable one.
How the score is built
- Attack vectorAn attacker can reach it over the internet; no access to your network needed.Network
- Attack complexityNo special conditions are needed; the attack works reliably.Low
- Privileges requiredThe attacker needs an ordinary user account.Low
- User interactionNo one has to click or open anything; the attack works on its own.None
- ScopeThe attack can spread beyond the vulnerable component to other parts of the system.Changed
- Confidentiality impactAn attacker can read all data the affected component holds, or the most sensitive parts of it.High
- Integrity impactAn attacker can change some data, but with limited control or limited effect.Low
- Availability impactThe component keeps working normally.None
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:L/A:NCISA has seen no public exploit and no attacks.
At least one step needs hands-on work, which slows mass attacks.
A successful attack gives limited control over the affected software or limited access to its data.
Public exploits
None foundNo exploit in Exploit-DB or Metasploit, and no scanner template. That lowers, but doesn't remove, the chance of mass exploitation.
Timeline
- 2025-11-10CVE publishedCVE List
- 2025-12-03F5 ships a ruleOAuth2-Proxy Header Smuggling
Vendor statements
QuotedCloudflareWAF Managed Rules
We found no public statement from this vendor naming this CVE. That is not proof of no protection: generic rules may still catch it. Each row gives the rule ID and its default action. The feed starts in January 2025, so older rules are missing.
AWSAWS WAF Managed Rules
We found no public statement from this vendor naming this CVE. That is not proof of no protection: generic rules may still catch it. AWS often describes a vulnerability without naming its CVE (for example "Log4j"). We count only entries that name the CVE, so AWS coverage is understated.
Microsoft AzureFront Door and Application Gateway WAF (DRS)
We found no public statement from this vendor naming this CVE. That is not proof of no protection: generic rules may still catch it. The rule tables name CVEs but carry no dates, so Azure is left out of every time-to-rule figure.
Google CloudCloud Armor preconfigured WAF rules
We found no public statement from this vendor naming this CVE. That is not proof of no protection: generic rules may still catch it. Google names few CVEs: the release notes start in May 2024 and the rules reference lists CVEs without dates. Its cve-canary rules are opt-in.
FastlyNext-Gen WAF virtual patches
The CVE predates the oldest entry in this vendor's public feed, so a rule may exist that we cannot see.
F5Attack signatures (Distributed Cloud WAF)
First rule 2025-12-03, 23 days after the CVE was published.
Added: signature 200013145 "OAuth2-Proxy Header Smuggling" (Authentication/Authorization Attacks, risk High), systems: PHP, Django, Python. References: CVE-2025-64484.
AkamaiApp & API Protector (Adaptive Security Engine)
This vendor publishes no machine-readable rule changelog we can check.
OWASP CRSCore Rule Set (open source, used inside many WAFs)
We found no public statement from this vendor naming this CVE. That is not proof of no protection: generic rules may still catch it. CRS rarely names CVEs: it relies on generic detections. Our own lab measures what it actually blocks.
Social
No posts seen yet on X, Bluesky, Mastodon, Hacker News or GitHub.