CVE-2026-79316: improper access control exposes xray

CISO Take

x-ui 0.3.2, a web panel used to manage xray proxy nodes, lets any authenticated panel user edit the xray configuration template and restart the panel, which rebinds the management gRPC service from loopback to non-loopback addresses. This turns a local-only management interface into a network-reachable one, and a low-privilege account is enough to trigger it. Exploitation likelihood is low for now: EPSS is 0.00322 (top 77th percentile), there is no public exploit, no Nuclei template, it is not in CISA KEV, and CISA SSVC rates it TRACK. The package is not part of the AI/ML packages we track and downstream dependents are unknown, so the AI blast radius is limited to organizations that use x-ui panels as proxy or egress infrastructure for AI workloads. Restrict panel accounts to trusted administrators, firewall the xray gRPC port from untrusted networks, and monitor for unexpected template changes and listener rebinds until a fixed release is available.

Sources:NVDEPSSATLASCISA SSVCGitHub (vaxilu/x-ui)

What is the risk?

Rated high (CVSS 7.6) with a network attack vector, low complexity, low privileges required and no user interaction. Confidentiality impact is high, integrity and availability impact are low. Real-world risk is moderated by the need for a valid panel account, the low EPSS (0.00322), no known public exploit and SSVC TRACK. Exposure rises sharply if the panel is internet-facing or shared with many users. The link to AI/ML is indirect: the product is a proxy management panel categorized as ml_ui, with no direct model or data pipeline involvement.

How does the attack unfold?

Initial Access

Attacker authenticates to the x-ui panel with a low-privilege account.

AML.T0012

Exploitation

Attacker modifies the xray configuration template through the settings interface and triggers a panel restart.

AML.T0049

Exposure

The xray management gRPC service is regenerated and bound to non-loopback addresses.

Impact

Attacker reaches the management API over the network to read or alter proxy configuration and disrupt traffic.

Initial Access

Attacker authenticates to the x-ui panel with a low-privilege account.

AML.T0012

Exploitation

Attacker modifies the xray configuration template through the settings interface and triggers a panel restart.

AML.T0049

Exposure

The xray management gRPC service is regenerated and bound to non-loopback addresses.

Impact

Attacker reaches the management API over the network to read or alter proxy configuration and disrupt traffic.

How severe is it?

CVSS 3.1

7.6 / 10

EPSS

0.3%

chance of exploitation in 30 days

Higher than 25% of all CVEs

Source: EPSS v3 — FIRST.org

Exploitation Status

No known exploitation

Sophistication

Trivial

What is the attack surface?

AVNetwork

ACLow

PRLow

UINone

SUnchanged

CHigh

ILow

ALow

What should I do?

1 step

1) Check the upstream repository (github.com/vaxilu/x-ui) for a fixed release and upgrade; if none exists, consider migrating to an actively maintained fork. 2) Limit panel accounts to trusted administrators and rotate their credentials. 3) Block the xray gRPC management port at the host firewall and security groups, and place the panel behind a VPN or IP allowlist. 4) Monitor for changes to the xray config template and unexpected panel restarts. 5) Scan for listeners on the xray API port bound to 0.0.0.0 or other non-loopback interfaces and alert on them.

What does CISA's SSVC say?

DecisionTrack

Exploitationnone

AutomatableNo

Technical Impactpartial

Source: CISA Vulnrichment (SSVC v2.0). Decision based on the CISA Coordinator decision tree.

How is it classified?

Auth BypassFrameworkAML.T0012- Valid Accounts AML.T0049- Exploit Public-Facing Application

Which compliance frameworks are affected?

This CVE is relevant to:

ISO 42001

A.6.2.6-AI system operation and monitoring

NIST AI RMF

MANAGE 2.3-Mechanisms to respond to and recover from unknown risks

Frequently Asked Questions

What is CVE-2026-79316?

x-ui 0.3.2, a web panel used to manage xray proxy nodes, lets any authenticated panel user edit the xray configuration template and restart the panel, which rebinds the management gRPC service from loopback to non-loopback addresses. This turns a local-only management interface into a network-reachable one, and a low-privilege account is enough to trigger it. Exploitation likelihood is low for now: EPSS is 0.00322 (top 77th percentile), there is no public exploit, no Nuclei template, it is not in CISA KEV, and CISA SSVC rates it TRACK. The package is not part of the AI/ML packages we track and downstream dependents are unknown, so the AI blast radius is limited to organizations that use x-ui panels as proxy or egress infrastructure for AI workloads. Restrict panel accounts to trusted administrators, firewall the xray gRPC port from untrusted networks, and monitor for unexpected template changes and listener rebinds until a fixed release is available.

Is CVE-2026-79316 actively exploited?

No confirmed active exploitation of CVE-2026-79316 has been reported, but organizations should still patch proactively.

How to fix CVE-2026-79316?

1) Check the upstream repository (github.com/vaxilu/x-ui) for a fixed release and upgrade; if none exists, consider migrating to an actively maintained fork. 2) Limit panel accounts to trusted administrators and rotate their credentials. 3) Block the xray gRPC management port at the host firewall and security groups, and place the panel behind a VPN or IP allowlist. 4) Monitor for changes to the xray config template and unexpected panel restarts. 5) Scan for listeners on the xray API port bound to 0.0.0.0 or other non-loopback interfaces and alert on them.

What systems are affected by CVE-2026-79316?

This vulnerability affects the following AI/ML architecture patterns: Proxy and egress gateways for ML infrastructure, Remote training and inference host access, Administrative web panels.

What is the CVSS score for CVE-2026-79316?

CVE-2026-79316 has a CVSS v3.1 base score of 7.6 (HIGH). The EPSS exploitation probability is 0.32%.

What is the AI security impact?

Affected AI Architectures

Proxy and egress gateways for ML infrastructureRemote training and inference host accessAdministrative web panels

MITRE ATLAS Techniques

AML.T0012Valid Accounts

AML.T0049Exploit Public-Facing Application

Compliance Controls Affected

ISO 42001: A.6.2.6

NIST AI RMF: MANAGE 2.3

What are the technical details?

Original Advisory

An improper access control vulnerability exists in x-ui 0.3.2. Any authenticated panel user can modify the xray configuration template through the settings interface and trigger a panel restart, causing the xray management gRPC service, which is bound to loopback by default, to be regenerated and bound to non-loopback addresses. This expands the reachable surface of the management interface beyond its intended local-only boundary.

Exploitation Scenario

An attacker obtains low-privilege credentials for an x-ui panel, for example through credential stuffing or a shared account, that fronts egress traffic for an ML team's remote training and inference hosts. They edit the xray configuration template so the gRPC API listens on a non-loopback address and trigger a panel restart. The management gRPC service is now reachable over the network, and they use it to read or alter proxy configuration, enumerate users and traffic, and disrupt or redirect connectivity to model-serving endpoints.

Weaknesses (CWE)

CWE-284Improper Access Control

CWE-284 — Improper Access Control: The product does not restrict or incorrectly restricts access to a resource from an unauthorized actor.

[Architecture and Design, Operation] Very carefully manage the setting, management, and handling of privileges. Explicitly manage trust zones in the software.
[Architecture and Design] Compartmentalize the system to have "safe" areas where trust boundaries can be unambiguously drawn. Do not allow sensitive data to go outside of the trust boundary and always be careful when interfacing with a compartment outside of the safe area. Ensure that appropriate compartmentalization is built into the system design, and the compartmentalization allows for and reinforces privilege separation functionality. Architects and designers should rely on the principle of least privilege to decide the appropriate time to use privileges and the time to drop privileges.

Source: MITRE CWE corpus.