CVE-2026-78608: missing authz leaks APM server

CISO Take

A missing authorization check on an internal Kibana APM integration function lets any authenticated user, regardless of privilege level, read APM server credentials that should be restricted to users with APM or Fleet administrative roles. This matters because the attack requires only network access and a low-privilege valid account (CVSS PR:L, AC:L, UI:N) — there is no social engineering or technical sophistication needed, and confidentiality impact is high. There's no evidence of active exploitation (not in CISA KEV, no EPSS score published, no public PoC or Nuclei template), so this looks like a disclosure caught before mass exploitation rather than an in-the-wild breach. Any organization running Elastic Stack with APM integration enabled to monitor application or AI/ML service performance should treat leaked APM credentials as a pivot point into observability data, which can include sensitive request/response traces. Patch to Kibana 8.19.21, 9.4.6, or 9.5.2 per the Elastic advisory, and rotate APM server credentials as a precaution since any authenticated user may have already retrieved them prior to patching.

Sources: NVD discuss.elastic.co

What is the risk?

Moderate risk overall (CVSS 6.5) but with a low exploitation bar: network-reachable, low attack complexity, no user interaction, and only low privileges required — meaning any authenticated Kibana account, even a low-tier one, can trigger it. Impact is confidentiality-only (no integrity or availability effect), limited to disclosure of APM server credentials rather than direct compromise of Kibana or the underlying data. Exposure scales with how many authenticated users a given Kibana instance has and whether APM integration is enabled; internet-facing Kibana deployments with broad user bases (common in larger SOC/observability teams) carry the highest exposure. No confirmed active exploitation, KEV listing, EPSS score, or public exploit/scanner signature currently exists, so urgency is driven by ease-of-exploitation rather than observed attacker activity.

How does the attack unfold?

Initial Access

Attacker obtains any valid, low-privileged authenticated Kibana user account.

AML.T0012

Authorization Bypass

Attacker invokes the internal Kibana APM integration function, which lacks an authorization check, to retrieve APM server credentials normally restricted to admin roles.

Credential Misuse / Impact

Attacker uses the leaked APM server credentials to access the APM server directly, reading trace/telemetry data or injecting spoofed data.

Initial Access

Attacker obtains any valid, low-privileged authenticated Kibana user account.

AML.T0012

Authorization Bypass

Attacker invokes the internal Kibana APM integration function, which lacks an authorization check, to retrieve APM server credentials normally restricted to admin roles.

Credential Misuse / Impact

Attacker uses the leaked APM server credentials to access the APM server directly, reading trace/telemetry data or injecting spoofed data.

How severe is it?

CVSS 3.1

6.5 / 10

EPSS

N/A

Exploitation Status

No known exploitation

Sophistication

Trivial

What is the attack surface?

AV Network

AC Low

PR Low

UI None

S Unchanged

C High

I None

A None

What should I do?

1 step

1) Patch Kibana to 8.19.21, 9.4.6, or 9.5.2 (or later) per the Elastic security advisory ESA-2026-144. 2) Rotate/regenerate APM server credentials immediately after patching, since any authenticated user could have retrieved them before the fix was applied. 3) Audit Kibana role assignments and remove unnecessary base-user access to the APM integration where feasible until patched. 4) Review APM server authentication logs for anomalous access patterns during the exposure window. 5) Restrict Kibana instance access to trusted networks/VPN where internet exposure isn't required, reducing the pool of accounts that could exploit this.

How is it classified?

Auth Bypass Data Leakage Plugin API AML.T0012 - Valid Accounts

Which compliance frameworks are affected?

This CVE is relevant to:

EU AI Act

Article 15 - Accuracy, robustness and cybersecurity

ISO 42001

A.6.2 - Information security for AI system operation

Frequently Asked Questions

What is CVE-2026-78608?

A missing authorization check on an internal Kibana APM integration function lets any authenticated user, regardless of privilege level, read APM server credentials that should be restricted to users with APM or Fleet administrative roles. This matters because the attack requires only network access and a low-privilege valid account (CVSS PR:L, AC:L, UI:N) — there is no social engineering or technical sophistication needed, and confidentiality impact is high. There's no evidence of active exploitation (not in CISA KEV, no EPSS score published, no public PoC or Nuclei template), so this looks like a disclosure caught before mass exploitation rather than an in-the-wild breach. Any organization running Elastic Stack with APM integration enabled to monitor application or AI/ML service performance should treat leaked APM credentials as a pivot point into observability data, which can include sensitive request/response traces. Patch to Kibana 8.19.21, 9.4.6, or 9.5.2 per the Elastic advisory, and rotate APM server credentials as a precaution since any authenticated user may have already retrieved them prior to patching.

Is CVE-2026-78608 actively exploited?

No confirmed active exploitation of CVE-2026-78608 has been reported, but organizations should still patch proactively.

How to fix CVE-2026-78608?

1) Patch Kibana to 8.19.21, 9.4.6, or 9.5.2 (or later) per the Elastic security advisory ESA-2026-144. 2) Rotate/regenerate APM server credentials immediately after patching, since any authenticated user could have retrieved them before the fix was applied. 3) Audit Kibana role assignments and remove unnecessary base-user access to the APM integration where feasible until patched. 4) Review APM server authentication logs for anomalous access patterns during the exposure window. 5) Restrict Kibana instance access to trusted networks/VPN where internet exposure isn't required, reducing the pool of accounts that could exploit this.

What systems are affected by CVE-2026-78608?

This vulnerability affects the following AI/ML architecture patterns: model serving, agent frameworks.

What is the CVSS score for CVE-2026-78608?

CVE-2026-78608 has a CVSS v3.1 base score of 6.5 (MEDIUM).

What is the AI security impact?

Affected AI Architectures

model servingagent frameworks

MITRE ATLAS Techniques

AML.T0012 Valid Accounts

Compliance Controls Affected

EU AI Act: Article 15

ISO 42001: A.6.2

What are the technical details?

Original Advisory

Missing Authorization (CWE-862) in Kibana can lead to information disclosure via Privilege Abuse (CAPEC-122). An authorization control was not applied to an internal Kibana APM integration function, allowing any authenticated Kibana user to read APM server credentials that should be restricted to users holding APM or Fleet administrative privileges.

Exploitation Scenario

An attacker obtains a low-privileged but valid authenticated Kibana account — for example through a phished credential, a shared/weak password, or an over-provisioned self-service signup in a larger organization. Without needing any special role like APM or Fleet admin, they invoke the vulnerable internal Kibana APM integration function directly (e.g., via its API endpoint) and retrieve the APM server credentials that should have been restricted. Using those credentials, the attacker connects directly to the APM server, where they can read ingested trace data — potentially including operational metadata about AI-powered services being monitored — or inject falsified telemetry to mask other malicious activity or mislead incident responders.

Weaknesses (CWE)

CWE-862 Missing Authorization

CWE-862 — Missing Authorization: The product does not perform an authorization check when an actor attempts to access a resource or perform an action.

[Architecture and Design] Divide the product into anonymous, normal, privileged, and administrative areas. Reduce the attack surface by carefully mapping roles with data and functionality. Use role-based access control (RBAC) [REF-229] to enforce the roles at the appropriate boundaries. Note that this approach may not protect against horizontal authorization, i.e., it will not protect a user from attacking others with the same role.
[Architecture and Design] Ensure that access control checks are performed related to the business logic. These checks may be different than the access control checks that are applied to more generic resources such as files, connections, processes, memory, and database records. For example, a database may restrict access for medical records to a specific database user, but each record might only be intended to be accessible to the patient and the patient's doctor [REF-7].

Source: MITRE CWE corpus.