CVERadar
HIGH · CVSS 3.1 8.1Public PoCEPSS 0.26% (p18)Threat level 45

CVE-2026-11800: CVE-2026-11800: A flaw was found in Keycloak. This JWT algorithm confusion vulnerability in the JWT Author…

redhat · build of keycloak · NVD published 2026-06-26 · posted 2026-11-10 · last modified 2026-10-26

CVE-2026-11800 at a glance

CVE-2026-11800 is a high-severity Signature Bypass vulnerability in Redhat Build Of Keycloak, with a CVSS 3.1 base score of 8.1. NVD published it on 2026-06-26 and classifies the weakness as CWE-347. Public exploit code: 3 public GitHub proof-of-concept repositories from 3 accounts, with the earliest exploit evidence dated 2026-11-09. FIRST EPSS puts the probability of exploitation activity in the next 30 days at 0.26% (percentile 18.2, scored 2026-11-10). NVD configuration data lists a fixed release for 1 product, for example redhat build of keycloak 26.6.4. CVERadar maps 2 MITRE ATT&CK techniques across Defense Impairment and Privilege Escalation, inferred from the CWE and the description. Its CVERadar threat level is 45/100.

What it is

A flaw was found in Keycloak. This JWT algorithm confusion vulnerability in the JWT Authorization Grant flow allows an attacker with valid client credentials to bypass signature verification. By forging an assertion, the attacker can create unauthorized access tokens. This enables the attacker to impersonate any federated user linked to the affected Identity Provider, leading to unauthorized access and potential privilege escalation.

WeaknessCWE-347 definition
CVSS vectorCVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N (source: Red Hat)
ATT&CK tacticsPrivilege Escalation → Defense Impairment

Exploitation evidence

First exploit evidence seen: 2026-11-09 (earliest date across the sources below).

Exploited in the wild

No source on this site lists in-the-wild exploitation yet (CISA KEV, VulnCheck KEV).

Public exploit code

GitHub proof-of-concept repositories

3 public repositories from 3 accounts, first seen 2026-11-09. Most starred: federiconbaez/cortex9-keycloak-jwt-grant-alg-confusion-model (0 stars), view the federiconbaez/cortex9-keycloak-jwt-grant-alg-confusion-model repository on GitHub.

Verify before running: public PoCs can be malicious. Read the code and only run it in an isolated lab. A PoC means exploit code is public, not that this CVE is exploited in the wild.

Sources: nomi-sec/PoC-in-GitHub.

Patches, mitigations and detection

Fixed versions

  • redhat build of keycloak: upgrade to 26.6.4 or later (affected from 26.6)

Derived from the NVD CPE configuration (versionEndExcluding / versionEndIncluding). Always confirm with the vendor advisory.

Vendor advisories and patches

MITRE ATT&CK mitigations for this chain

  • M1051 Update Software (for T1068) : Update software regularly by employing patch management for internal enterprise endpoints and servers.
  • M1050 Exploit Protection (for T1068) : Security applications that look for behavior used during exploitation such as Windows Defender Exploit Guard (WDEG) and the Enhanced Mitigation Experience Toolkit (EMET) can be used to mitigate some exploitation behavior. Control flow integrity checking is another way to potentially identify and stop a software exploi
  • M1048 Application Isolation and Sandboxing (for T1068) : Make it difficult for adversaries to advance their operation through exploitation of undiscovered or unpatched vulnerabilities by using sandboxing. Other types of virtualization and application microsegmentation may also mitigate the impact of some types of exploitation. Risks of additional exploits and weaknesses in t
  • M1019 Threat Intelligence Program (for T1068) : Develop a robust cyber threat intelligence capability to determine what types and levels of threat may use software exploits and 0-days against a particular organization.
  • M1038 Execution Prevention (for T1068, T1553) : Consider blocking the execution of known vulnerable drivers that adversaries may exploit to execute code in kernel mode. Validate driver block rules in audit mode to ensure stability prior to production deployment.
  • M1028 Operating System Configuration (for T1553) : Windows Group Policy can be used to manage root certificates and the <code>Flags</code> value of <code>HKLM\\SOFTWARE\\Policies\\Microsoft\\SystemCertificates\\Root\\ProtectedRoots</code> can be set to 1 to prevent non-administrator users from making further root installations into their own HKCU certificate store.
  • M1026 Privileged Account Management (for T1553) : Manage the creation, modification, use, and permissions associated to privileged accounts, including SYSTEM and root.
  • M1024 Restrict Registry Permissions (for T1553) : Ensure proper permissions are set for Registry hives to prevent users from modifying keys related to SIP and trust provider components. Components may still be able to be hijacked to suitable functions already present on disk if malicious modifications to Registry keys are not prevented.
  • M1054 Software Configuration (for T1553) : HTTP Public Key Pinning (HPKP) is one method to mitigate potential [Adversary-in-the-Middle](https://attack.mitre.org/techniques/T1557) situations where and adversary uses a mis-issued or fraudulent certificate to intercept encrypted communications by enforcing use of an expected certificate.

Detection Generic guidance

  • Privilege Escalation: Alert on processes gaining SYSTEM/root unexpectedly, and on exploitation crashes of privileged services.

Response checklist Generic guidance

  1. Inventory: find every instance of the affected product and version (include internet-facing and forgotten ones).
  2. Patch or mitigate per the vendor advisory; if you cannot patch, remove internet exposure or apply the vendor workaround.
  3. Assume compromise if the system was exposed before patching: hunt for the indicators in the advisory and review logs back to the disclosure date.
  4. Rotate credentials and secrets stored on or used by the affected system if compromise cannot be ruled out.

If an attacker gets into Redhat Build Of Keycloak, they get into everything behind it; identity monitoring from LayerLogix looks for that.

Sources