Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
ReconRegister malicious OIDC client in target realm
-
DeliveryPhish victim with crafted authentication link
-
ExploitVictim authenticates through attacker's client
-
InstallPredict account-linking hash for victim session
-
C2Submit forged linking URL during victim's session
-
ExecuteKeycloak binds victim account to attacker's external identity
-
ImpactAttacker logs in as victim
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A04:2025 - Cryptographic Failures
ASVS 5.0
V11 - Cryptography Predictable value prevention
|
| Exploitation Conditions | Three concrete prerequisites must be met simultaneously. First, the legacy client-initiated account-linking endpoint must be active in the Keycloak deployment - this is explicitly described as a 'legacy' mechanism, so newer or hardened deployments that have disabled or removed this endpoint are not affected. Second, the attacker must control a registered OIDC client within the target Keycloak realm; the CVSS PR:L rating reflects this requirement - if client self-registration is disabled and the attacker has no realm access, they cannot register such a client and exploitation is blocked. Third, the victim must be actively deceived into initiating an authentication session through the attacker's client (CVSS UI:R), requiring social engineering or phishing. Exploitation is materially limited by: the user-interaction requirement ruling out automated/mass exploitation, the need for a registered client in the specific target realm, and the legacy nature of the vulnerable endpoint which may already be absent in modern deployments. |
| Impact | The CVSS 3.1 vector (AV:N/AC:L/PR:L/UI:R/S:U/C:H/I:H/A:N) scores 7.3 High, which is broadly consistent with the described impact. AV:N and AC:L indicate the attack is network-accessible and requires no special circumstances beyond controlling a registered client - an important prerequisite encoded in PR:L. The UI:R requirement is the primary limiting factor: exploitation depends on a victim being socially engineered into clicking a phishing link and completing an authentication flow, making mass automated exploitation unlikely. No EPSS data was provided, and the absence of a KEV listing means active exploitation is not confirmed. The account-takeover outcome (C:H/I:H) is severe and there is no scope change (S:U) since the attacker's impact is bounded to the victim's identity within the realm. Overall, real-world risk is moderate-to-high for organizations where: (a) the legacy endpoint is active, (b) client registration is open or the attacker has prior realm access, and (c) users can be targeted via phishing. For deployments with strict client registration controls, risk is meaningfully lower. |
| Exploit ScenarioDerived | An attacker registers a malicious OIDC client in the target Keycloak realm (possible when client registration is unrestricted or the attacker has obtained low-level realm access), then sends the victim a phishing link that initiates an authentication flow through the attacker's client. By predicting the hash value used in the legacy account-linking protection mechanism, the attacker constructs a forged linking URL; once the victim authenticates, Keycloak processes the forged URL and permanently links the victim's account to the attacker's external identity, after which the attacker logs in as the victim at will. No public proof-of-concept has been identified at time of analysis. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Aug 18, 2026
Full account takeover is possible in Keycloak's legacy client-initiated account-linking endpoint due to a predictable hash (CWE-341) used as a CSRF-like protection mechanism. An attacker who controls a registered OIDC client in the target realm can forge a valid linking URL that silently binds a victim's Keycloak account to the attacker's external identity, enabling subsequent login as the victim. No public exploit or CISA KEV listing is identified at time of analysis; Red Hat has released patches via RHSA-2026:56523 and RHSA-2026:56524.
RemediationAI
Apply Red Hat patches RHSA-2026:56523 and RHSA-2026:56524 (https://access.redhat.com/errata/RHSA-2026:56523 and https://access.redhat.com/errata/RHSA-2026:56524); the exact patched Keycloak version number is not independently confirmed from the available data and should be verified against the errata details. If immediate patching is not feasible, disable the legacy client-initiated account-linking endpoint to eliminate the attack surface - this will disable client-initiated account federation but does not affect standard authentication flows; administrators should assess the functional impact in their environment before applying this workaround. Restrict OIDC client registration to prevent attackers from registering malicious clients; this raises the barrier for exploitation from PR:L toward a higher access requirement. Audit existing account links in the Keycloak admin console for unexpected or unauthorized external identity bindings and revoke any suspicious associations.
Affected VersionsENISA
Affects your stack
Find out whether the stacks and servers you run are on an affected version.
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
Keycloak supports account federation, allowing users to link their local account to external identity providers through OIDC. The legacy client-initiated account-linking flow protects against unauthorized link requests using a hash-based token. This hash is generated using a predictable algorithm (CWE-341: Predictable Pseudo-Random Number Generator), meaning an attacker who controls a registered OIDC client can calculate or brute-force the expected hash value for a given victim session and forge a complete, valid account-linking URL. The forged URL, when followed by the victim during an attacker-triggered authentication flow, causes Keycloak to associate the victim's account with the attacker's external identity. No CPE strings were provided in the input data, but the vulnerability is described as residing in the legacy endpoint, implying impact on all Keycloak versions that retain this code path.
Know when the next one hits your stack
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
AV:N/AC:L/PR:L/UI:R/S:U/C:H/I:H/A:N
Network endpoint, predictable hash means AC:L; attacker needs a registered OIDC client (PR:L); victim must authenticate through attacker's client (UI:R); no availability impact.
Primary rating from Vendor (redhat).
CVSS VectorVendor: redhat
DescriptionCVE.org
A flaw was found in the legacy client-initiated account-linking endpoint of Keycloak, a widely used open-source identity and access management solution. The mechanism used to protect the account-linking process from unauthorized requests relies on a hash that can be predicted by a malicious OIDC client. By tricking a user into authenticating, an attacker-controlled client can forge a valid linking URL to connect the victim's account to an attacker's external identity. This results in a full account takeover, allowing the attacker to log in as the victim.
Executive ViewAI
Keycloak is vulnerable to full account takeover through its legacy account-linking mechanism due to a cryptographically weak CSRF protection implementation (CVE-2026-15571, CVSS 7.3). Any attacker with a registered OIDC client in the target realm can silently hijack victim user accounts without authentication, directly compromising identity infrastructure and enabling unauthorized access to all dependent systems.
Apply vendor patch.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Authentication bypass attacks exploit flaws in the verification mechanisms that control access to systems and applications.
EUVD-2026-61238
GHSA-m639-f4cw-m3hm
NVD
CVE.org
EPSS
Exploit-DB