What is the CVSS score of CVE-2026-63145?

CVE-2026-63145 has a CVSS 3.1 base score of 4.3 (Medium). CVSS vector: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:L/A:N. EPSS exploitation probability: 0.2%.

Kibana CVE-2026-63145

EUVD-2026-47584 MEDIUM

Incorrect Authorization (CWE-863)

2026-07-21 security@elastic.co

GHSA-q5gv-xxq5-3m7c

Elastic Authentication Bypass Kibana

4.3

CVSS 3.1 · Vendor: elastic

Severity by source

Vendor (elastic) PRIMARY

4.3 MEDIUM

AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:L/A:N

vuln.today AI

4.3 MEDIUM

PR:L because ML access is required; I:L because only ML audit/notification record integrity is affected; C:N and A:N because no data read or service disruption occurs.

3.1 AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:L/A:N

4.0 AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:L/VA:N/SC:N/SI:N/SA:N

Primary rating from Vendor (elastic).

CVSS VectorVendor: elastic

Attack Vector

Network

Attack Complexity

Low

Privileges Required

Low

User Interaction

None

Scope

Unchanged

Confidentiality

None

Integrity

Low

Availability

None

Lifecycle Timeline

Analysis Generated

Jul 22, 2026 - 07:59 vuln.today

CVE Published

Jul 21, 2026 - 23:18 cve.org

MEDIUM 4.3

DescriptionCVE.org

Incorrect Authorization (CWE-863) in Kibana can lead to integrity compromise of Machine Learning audit and notification records via Accessing Functionality Not Properly Constrained by ACLs (CAPEC-1).

A vulnerability exists in Kibana's Machine Learning functionality where a Machine Learning management endpoint performs an insufficient authorization check. The endpoint validates only a coarse privilege level but does not verify that the requesting user has access to the specific Machine Learning job or notification resources provided in the request. As a result, a low-privileged user with Machine Learning access in any Kibana space can manipulate Machine Learning audit and notification records for arbitrary jobs-including jobs in other spaces or belonging to other users-by leveraging Kibana's internally elevated credentials to write to restricted Machine Learning system indices that the user cannot access directly.

AnalysisAI

Incorrect Authorization in Kibana's Machine Learning management endpoint allows a low-privileged authenticated user to manipulate ML audit and notification records for arbitrary jobs across any space or user boundary. The endpoint validates only a coarse ML access privilege without enforcing resource-level authorization, and then uses Kibana's internally elevated credentials to write to restricted ML system indices on behalf of the requesting user. No public exploit has been identified at time of analysis, and CVSS scores this at Medium (4.3) given the limited integrity-only impact scope.

Technical ContextAI

Kibana is the visualization and management front-end for the Elastic Stack. Its Machine Learning functionality exposes management endpoints that interact with Elasticsearch's restricted ML system indices. The root cause is CWE-863 (Incorrect Authorization): the endpoint performs a coarse privilege check - confirming only that the user has some ML access - but omits object-level authorization to verify whether the user is permitted to act on the specific job or notification resource referenced in the request. This is consistent with CAPEC-1 (Accessing Functionality Not Properly Constrained by ACLs). Critically, Kibana performs the actual Elasticsearch write using its own internally elevated service credentials, effectively laundering the unprivileged user's request through a trusted identity. The affected versions span the 8.x and 9.x branches as specified by EUVD-2026-47584.

RemediationAI

Upgrade to a fixed release: Kibana 8.19.19, Kibana 9.3.8, or Kibana 9.4.4, as announced in Elastic security advisory ESA-2026-69 at https://discuss.elastic.co/t/kibana-8-19-19-9-3-8-9-4-4-security-update-esa-2026-69/388572. These are the minimum fixed versions inferred from the affected version ranges provided by EUVD-2026-47584; independently verify exact patch version with Elastic's release notes. If immediate patching is not possible, a targeted compensating control is to restrict the Machine Learning privilege to only fully trusted users in each Kibana space, minimizing the population of users who can reach the vulnerable endpoint - note this may impact legitimate ML workflows. Removing ML access from users who do not operationally require it reduces exposure without disabling the feature entirely. No generic network-layer workaround is effective since the flaw is in application-layer authorization logic.