CVE-2026-4399: Software Security Vulnerability Vulnerability

HIGH
Advisory Identifier: GHSA-7649-76jc-fc5r
Share
CVSS 7.5
CWE-77
Published
Advisory Summary

CVE-2026-4399 is a HIGH (CVSS 7.5) security vulnerability vulnerability (CWE-77: Improper Neutralization of Special Elements used in a Command ('Command Injection')) affecting Software libraries in the Maven ecosystem. The weakness occurs when the product constructs all or part of a command using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the intended command when it is sent to a downstream component.. Remediation is available in the latest vendor maintenance releases.

CVE-2026-4399 at a Glance

CVE Identifier CVE-2026-4399
GHSA Identifier GHSA-7649-76jc-fc5r
Severity HIGH
CVSS v3.1 Base Score 7.5
CWE Classification CWE-77: Improper Neutralization of Special Elements used in a Command ('Command Injection')
Affected Product Software
Package Ecosystem Maven
Published Date
Last Updated

Security Impact

CVSS: 7.5

The security impact metrics below are parsed directly from the CVSS v3.1 vector string (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:H/A:N).

Attack Vector Network (AV:N)
Attack Complexity Low (AC:L)
Privileges Required None (PR:N)
User Interaction None (UI:N)
Scope Unchanged (S:U)
Confidentiality None (C:N)
Integrity High (I:H)
Availability None (A:N)

How to Fix CVE-2026-4399

To remediate this vulnerability, upgrade all affected Software dependencies to a patched release. Both direct and transitive dependencies should be checked, as frameworks or security libraries may pull in vulnerable versions transitively.

Verify Direct and Transitive Dependencies

Run the following command in your project directory to verify that all resolved dependencies use the patched release: 

mvn dependency:tree

How to Check if You Are Affected

To determine whether your Java application is affected by CVE-2026-4399, inspect your build dependencies using the Maven Dependency Plugin:

mvn dependency:tree

Analyze the terminal output to locate any version of Software in your dependency tree. If any resolved version matches the vulnerable ranges listed in the table below (for example, versions earlier than the patched release), your application is affected.

Note: The mvn dependency:tree command inspects the resolved dependency graph declared in your project's POM files; it does not perform automated security scanning or inspect runtime dynamic classpaths. Always verify runtime container classpaths and shaded JARs.

Technical Details

Vulnerability Mechanism & Technical Analysis

This vulnerability is classified under CWE-77: Improper Neutralization of Special Elements used in a Command ('Command Injection'), where the product constructs all or part of a command using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the intended command when it is sent to a downstream component. In Software, software state validation or input boundaries are circumvented during processing. Review the official source advisory statement below for specific affected components, classes, and reproduction details.

Source Advisory Statement

Prompt injection vulnerability in 1millionbot Millie chatbot that occurs when a user manages to evade chat restrictions using Boolean prompt injection techniques (formulating a question in such a way that, upon receiving an affirmative response ('true'), the model executes the injected instruction), causing it to return prohibited information and information outside its intended context. Successful exploitation of this vulnerability could allow a malicious remote attacker to abuse the service for purposes other than those originally intended, or even execute out-of-context tasks using 1millionbot's resources and/or OpenAI's API key. This allows the attacker to evade the containment mechanisms implemented during LLM model training and obtain responses or chat behaviors that were originally restricted.

References

NVD — CVE-2026-4399 NVD
https://nvd.nist.gov/vuln/detail/CVE-2026-4399
www.incibe.es/en/incibe-cert/noti... WEB
https://www.incibe.es/en/incibe-cert/notices/aviso/multiple-vulnerabilities-1millionbot-millie-chatbot

Frequently Asked Questions

What is CVE-2026-4399?
CVE-2026-4399 is a security vulnerability in Software classified as a security vulnerability vulnerability under CWE-77: Improper Neutralization of Special Elements used in a Command ('Command Injection').
Which Software versions are affected?
Affected versions include releases prior to current release patches.
What version fixes CVE-2026-4399?
Upgrading to current release patches (or the latest patched release for your respective branch) resolves the vulnerability.
What is the CVSS score of CVE-2026-4399?
CVE-2026-4399 has a CVSS v3.1 base score of 7.5 (HIGH) with vector string CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:H/A:N.
What CWE is associated with CVE-2026-4399?
CVE-2026-4399 is classified under CWE-77: Improper Neutralization of Special Elements used in a Command ('Command Injection') (The product constructs all or part of a command using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the intended command when it is sent to a downstream component.).
How can I check Maven dependencies for CVE-2026-4399?
Run the dependency tree command: 'mvn dependency:tree'. This command lists direct and transitive dependencies in your build configuration so you can verify if a vulnerable version is resolved.
Machine-readable OSV JSON OSV v1.5 Schema
Normalized schema output
View Raw JSON 
{
  "schema_version" : "1.4.0",
  "id" : "GHSA-7649-76jc-fc5r",
  "modified" : "2026-04-13T15:31:07Z",
  "published" : "2026-03-31T12:31:35Z",
  "aliases" : [ "CVE-2026-4399" ],
  "details" : "Prompt injection vulnerability in 1millionbot Millie chatbot that occurs when a user manages to evade chat restrictions using Boolean prompt injection techniques (formulating a question in such a way that, upon receiving an affirmative response ('true'), the model executes the injected instruction), causing it to return prohibited information and information outside its intended context. Successful exploitation of this vulnerability could allow a malicious remote attacker to abuse the service for purposes other than those originally intended, or even execute out-of-context tasks using 1millionbot's resources and/or OpenAI's API key. This allows the attacker to evade the containment mechanisms implemented during LLM model training and obtain responses or chat behaviors that were originally restricted.",
  "severity" : [ {
    "type" : "CVSS_V3",
    "score" : "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:H/A:N"
  }, {
    "type" : "CVSS_V4",
    "score" : "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X"
  } ],
  "affected" : [ ],
  "references" : [ {
    "type" : "ADVISORY",
    "url" : "https://nvd.nist.gov/vuln/detail/CVE-2026-4399"
  }, {
    "type" : "WEB",
    "url" : "https://www.incibe.es/en/incibe-cert/notices/aviso/multiple-vulnerabilities-1millionbot-millie-chatbot"
  } ],
  "database_specific" : {
    "cwe_ids" : [ "CWE-77" ],
    "severity" : "HIGH",
    "github_reviewed" : false,
    "github_reviewed_at" : null,
    "nvd_published_at" : "2026-03-31T11:16:14Z"
  }
}