Skip to content
SentinelOne Vulnerability Database: CVE-2026-39118

SentinelOne Vulnerability Database: CVE-2026-39118

www.sentinelone.com • June 25, 2026

CVE-2026-39118 is a privilege escalation vulnerability in the Iru, Inc Kandji Agent prior to version 4.7.5(5374) . The flaw stems from a client-side validation gap that allows a local attacker to invoke restricted agent functionality reserved for privileged operations. Kandji Agent runs on managed macOS endpoints with elevated permissions, making any local bypass a direct path to higher privileges. The issue is categorized under [CWE-269: Improper Privilege Management].

A local, unauthenticated attacker can escalate privileges on affected endpoints and execute restricted agent operations, leading to full compromise of confidentiality, integrity, and availability.

The Kandji Agent exposes privileged functionality intended to be invoked only by trusted, authenticated components. Versions prior to 4.7.5(5374) enforce these restrictions on the client side rather than within the privileged service boundary. A local attacker who can communicate with the agent can bypass these checks and trigger restricted operations directly.

Because the agent operates with elevated privileges on macOS hosts, abuse of these interfaces translates into immediate privilege escalation. The advisory does not disclose proof-of-concept code or specific interface names, but the attack requires only local access with no prior authentication and no user interaction.

The root cause is improper privilege management ([CWE-269]). Authorization decisions are made by the calling client rather than enforced by the agent's privileged service. Any local process able to construct valid-looking requests to the agent can therefore invoke functionality that should require elevated authorization.

Exploitation requires local access to a host running a vulnerable Kandji Agent. An attacker with standard user privileges interacts with the agent's local interface and issues requests for restricted operations. Without server-side validation, the agent executes the operation in its privileged context, granting the attacker the agent's effective permissions.

The vulnerability mechanism is described in prose only; no verified public exploit code is available. Refer to the IRU Update on Agent 4.7.5 for vendor-supplied technical context.

Iru, Inc addressed the vulnerability in Kandji Agent 4.7.5(5374) . Deployment details and release notes are available in the IRU Update on Agent 4.7.5 . Administrators should validate that all managed endpoints report the patched build through the Kandji console.

Disclaimer : This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.

Extracted Entities