Jenkins 2.587 and earlier, LTS 2.580.1 and earlier improperly identifies file paths attempting path traversal in the file browser for workspaces and $JENKINSHOME/userContent/, allowing attackers with Overall/Read permission to obtain the names, sizes, and last modification times of files in the $JENKINSHOME, and attackers with Item/Workspace permission to obtain the names of jobs that have a workspace on the same agent.

A missing permission check in Jenkins 2.587 and earlier, LTS 2.580.1 and earlier allows attackers with Overall/Read permission to obtain the server, port, user name, and no-proxy host list of the configured HTTP proxy.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier does not consistently enforce the restriction on deserializing objects of types marked as storing their configuration in independent top-level configuration files as nested values, allowing attackers to bypass the root element check and potentially inject nested objects.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier does not escape the text shown in the tooltips of breadcrumb bar overflow menu entries, resulting in a stored cross-site scripting (XSS) vulnerability exploitable by attackers with Overall/Read permission.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier computes on the agent the URL of the link shown in the workspace browser for each file and directory, resulting in a cross-site scripting (XSS) vulnerability exploitable by attackers in control of agent processes.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier allows agents to send exceptions implementing the Stapler HttpResponse interface to the controller, resulting in a cross-site scripting (XSS) vulnerability exploitable by attackers in control of agent processes.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier allows requests authenticated using HTTP Basic authentication, including with an API token, to obtain a "remember me" cookie, allowing attackers who have obtained a user's API token to retain access to Jenkins using a "remember me" cookie after the API token is revoked or expires.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier allows attackers with Overall/Read permission to trigger warnings in the "Manage Old Data" feature that can be acted on by administrators to inject attacker-controlled data, overriding existing user account information, including login passwords.

In Jenkins 2.587 and earlier, LTS 2.580.1 and earlier, the HTTP endpoint for creating the first user account does not check whether the setup wizard has been completed, resulting in a race condition that allows unauthenticated attackers to create an administrator account while the initial user account is being replaced.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier checks for symbolic links in the path of an extracted file only before creating its parent directories, not immediately before writing it, and allows symbolic links in artifacts to replace existing files or empty directories, allowing attackers with Item/Configure permission to write files to arbitrary locations on the Jenkins controller file system.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier provides an HTTP endpoint that loads agents by name without validating the name, allowing attackers with Overall/Read permission to load any config.xml file on the controller file system as an agent configuration, and to overwrite an arbitrary config.xml file when that agent is saved, potentially resulting in arbitrary code execution.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier follows symbolic links when copying the directory of an item being renamed, allowing attackers with Item/Create and Item/Configure permission to read arbitrary files on the Jenkins controller file system.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier converts the launch commands in older agent configurations into launchers provided by the Command Agent Launcher Plugin without checking whether they have been approved by an administrator, allowing attackers with Agent/Configure permission to execute arbitrary code in the context of the Jenkins controller JVM.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier does not restrict item types in collections after deserialization, allowing attackers with access to POST config.xml API endpoints to submit crafted config.xml files that can lead to, e.g., stored cross-site scripting (XSS) vulnerabilities.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier does not apply its deserialization filter to classes referenced by the resolves-to attribute during XStream deserialization.

Jenkins is an open source automation server which enables developers around the world to reliably build, test, and deploy their software.

The following releases contain fixes for security vulnerabilities:

Jenkins 2.588 Jenkins LTS 2.580.2

Summaries of the vulnerabilities are below. More details, severity, and attribution can be found here: https://www.jenkins.io/security/advisory/2026-10-28/

We provide advance notification for security updates on this mailing list: https://groups.google.com/d/forum/jenkinsci-advisories

If you discover security vulnerabilities in Jenkins, please report them as described here: https://www.jenkins.io/security/#reporting-vulnerabilities

---

SECURITY-4125 / CVE-2026-117612 Jenkins uses XStream for serialization and deserialization of configuration and build data. To protect from common deserialization vulnerabilities, Jenkins uses a custom deserialization filter that only allows deserialization of types defined in Jenkins core or plugins, or explicitly allowed types (JEP-200). These vulnerabilities generally rely on behavior defined in #readResolve methods that are executed during deserialization.

The custom XStream converter used by Jenkins 2.587 and earlier, LTS 2.580.1 and earlier does not apply the deserialization filter to classes referenced by the resolves-to attribute. This allows attackers to instantiate arbitrary types during deserialization, potentially leading to remote code execution. To do this, attackers need to have Overall/Read permission, and at least one of the following:

a user account (i.e., cannot be anonymous) any set of permissions allowing them to POST config.xml (e.g., Item/Configure, View/Configure, Agent/Configure).

SECURITY-4225 / CVE-2026-117613 Jenkins uses XStream for serialization and deserialization of configuration and build data. The ability to provide XML documents to XStream is exposed to users who hold the respective /Configure permission, e.g., the item POST config.xml API is available with Item/Configure permission.

XStream by itself does not restrict the types of collection elements after deserialization. While some converters in Jenkins restrict collection element types (see SECURITY-3707 in the 2026-06-10 security advisory]), not all converters restrict element types.

In Jenkins 2.587 and earlier, LTS 2.580.1 and earlier this allows attackers with Item/Configure permission to submit a crafted job config.xml file that causes a stored cross-site scripting (XSS) vulnerability.

TIP: On Jenkins 2.539 and newer, LTS 2.541.1 and newer, enforcing Content Security Policy protection mitigates the known impact of this vulnerability.

Other XSS vulnerabilities, or other vulnerability types stemming from unexpected collection element types, may also be possible.

SECURITY-4123 / CVE-2026-117614 Since Jenkins 2.86, LTS 2.89.1 launching agents by running a command on the controller is implemented in the Command Agent Launcher Plugin rather than in Jenkins.

For backward compatibility with older agent configurations, Jenkins 2.587 and earlier, LTS 2.580.1 and earlier converts the launch commands they contain into launchers provided by this plugin.

This conversion does not check whether the launch command has been approved by an administrator, allowing attackers with Agent/Configure permission to execute arbitrary code in the context of the Jenkins controller JVM.

SECURITY-4187 / CVE-2026-117615 When renaming an item, Jenkins moves its directory on the controller file system, or copies it if it cannot be moved.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier follows symbolic links when copying the directory of an item being renamed.

This allows attackers with Item/Create and Item/Configure permission to read arbitrary files on the Jenkins controller file system, which can lead to remote code execution.

SECURITY-4188 / CVE-2026-117616 Jenkins 2.587 and earlier, LTS 2.580.1 and earlier provides an HTTP endpoint that loads agents by name without validating the name, so that any config.xml file on the controller file system can be loaded, as an agent carrying the name declared in that file.

This allows attackers with Overall/Read permission to load any config.xml file on the controller file system as an agent configuration, replacing the configuration of an existing agent or defining a new agent. The name declared in that file also determines where the agent's configuration is written. When this agent is saved, Jenkins overwrites the config.xml file at that location, potentially resulting in arbitrary code execution.

SECURITY-4248 / CVE-2026-117617 To prevent writing files outside the target directory, Jenkins refuses to write files through symbolic links when extracting archived artifacts on the controller.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier checks for symbolic links in the path of an extracted file only before creating its parent directories, not immediately before writing it, and allows symbolic links in artifacts to replace existing files or empty directories.

This allows attackers with Item/Configure permission to write files to arbitrary locations on the Jenkins controller file system by archiving artifacts concurrently, e.g., from parallel branches of a Pipeline, which can lead to remote code execution.

SECURITY-4232 / CVE-2026-117621 When the workspace of a job is located on an agent, Jenkins retrieves the list of files and directories shown in the workspace browser from that agent.

In Jenkins 2.587 and earlier, LTS 2.580.1 and earlier, the agent computes the URL of the link shown for each file and directory.

This results in a cross-site scripting (XSS) vulnerability exploitable by attackers in control of agent processes.

TIP: On Jenkins 2.539 and newer, LTS 2.541.1 and newer, enforcing Content Security Policy protection mitigates this vulnerability.

SECURITY-4249 / CVE-2026-117622 To protect against deserialization vulnerabilities, Jenkins restricts the types of objects that agents can send to the controller (JEP-200).

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier allows agents to send exceptions implementing the Stapler HttpResponse interface, and renders them as HTTP responses.

This results in a cross-site scripting (XSS) vulnerability exploitable by attackers in control of agent processes.

SECURITY-4229 / CVE-2026-117623 When the breadcrumb bar does not fit in the browser window, Jenkins moves breadcrumbs into an overflow menu, and shows the full text of long breadcrumbs, such as job, build, or user names, in a tooltip.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier does not escape the text shown in these tooltips.

This results in a stored cross-site scripting (XSS) vulnerability exploitable by attackers with Overall/Read permission.

TIP: On Jenkins 2.539 and newer, LTS 2.541.1 and newer, enforcing Content Security Policy protection mitigates this vulnerability.

SECURITY-4122 / CVE-2026-117618 By default, the initial setup of Jenkins configures the "Jenkins' own user database" security realm and the "Logged-in users can do anything" authorization strategy, and creates an initial user account. During the setup wizard, administrators can create a new administrator account, in which case Jenkins deletes the initial account and then creates the new account.

To prevent administrators from being locked out, the "Jenkins' own user database" security realm provides an HTTP endpoint that allows unauthenticated users to create a user account while no user accounts exist.

In Jenkins 2.587 and earlier, LTS 2.580.1 and earlier, this HTTP endpoint does not check whether the setup wizard has been completed, resulting in a race condition that allows unauthenticated attackers to create an administrator account while the initial account is being replaced.

SECURITY-4189 / CVE-2026-117619 Jenkins's "Keep me signed in" feature uses a persistent cookie, known as the "remember me" cookie, to re-authenticate users without requiring them to enter credentials again. The validity of this cookie does not depend on the credentials used to obtain it.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier allows requests authenticated using HTTP Basic authentication, including with an API token, to obtain a "remember me" cookie.

This allows attackers who have obtained a user's API token to retain access to Jenkins using a "remember me" cookie after the API token is revoked or expires.

SECURITY-4226 / CVE-2026-117620 Jenkins uses XStream for serialization and deserialization of configuration and build data. The ability to provide XML documents to XStream is exposed to users who hold the respective /Configure permission, e.g., the item POST config.xml API is available with Item/Configure permission.

When invalid data (e.g., data for a field that no longer exists) is loaded, this is recorded by the "Manage Old Data" feature if the data is being loaded from disk to allow administrators to review and correct any errors. It also allows saving the invalid objects again to remove any stored obsolete data.

In Jenkins 2.587 and earlier, LTS 2.580.1 and earlier, some of the code paths to report invalid data are not restricted to data being loaded from disk. This allows attackers with Overall/Read permission to trigger a warning in the "Manage Old Data" feature. When an administrator acts on the warning, it can override a Jenkins user account's data with attacker-controlled content. On controllers using Jenkins's own user database, this includes the password for that user account, so that the attacker can log in as that user afterwards.

SECURITY-4243 / CVE-2026-117624 Jenkins uses XStream for serialization and deserialization of configuration and build data. Types storing their configuration in their own files (usually config.xml or build.xml), like agents, items, or builds, are indicated as such by implementing the Java interface PersistenceRoot.

Since SECURITY-3972 in the 2026-09-02 security advisory, Jenkins attempts to prevent objects of types marked as storing their configuration in independent top-level configuration files in Jenkins from being deserialized as nested values in the configuration of other objects.

Jenkins 2.587 and earlier, LTS 2.580.1 and earlier does not consistently enforce this restriction, allowing attackers to bypass the root element check and potentially inject malicious nested objects.

As of publication, the confirmed impact of this vulnerability is limited to the creation of new user records (name and display name only), similar to users appearing in SCM logs. Notably, no new user accounts with the ability to log in can be created (unlike in SECURITY-4226).

SECURITY-4206 / CVE-2026-117625 Jenkins 2.587 and earlier, LTS 2.580.1 and earlier does not perform a permission check in an HTTP endpoint.

This allows attackers with Overall/Read permission to obtain the server, port, user name, and no-proxy host list of the configured HTTP proxy.

SECURITY-4237 / CVE-2026-117626 Jenkins 2.587 and earlier, LTS 2.580.1 and earlier improperly identifies file paths attempting path traversal in the file browser for workspaces and $JENKINSHOME/userContent/.

This allows attackers with Overall/Read permission to obtain the names, sizes, and last modification times of files in $JENKINSHOME. Additionally, it allows attackers with Item/Workspace permission to obtain the names of jobs that have a workspace on the same agent.

-- CONFIDENTIALITY NOTICE: This email and any attachments contain confidential and proprietary information of CloudBees intended only for the named recipient(s). Unauthorized use or distribution is prohibited. If you received this in error, please notify the sender and delete this email.

Jenkins Keycloak Authentication Plugin 2.4.1 and earlier does not restrict the redirect URL after login, allowing attackers to perform phishing attacks.

The OAuth authorization endpoint in Jenkins Bitbucket Server Integration Plugin 6.0.1 and earlier reads the oauthcallback URL from the submitted form rather than from the server-side stored request token, allowing attackers to hijack the OAuth flow and obtain an access token on behalf of the victim.

Jenkins Robot Framework Plugin 6.2.2 and earlier does not check that the archive directory configured for Robot Framework report files is contained within the build directory on the Jenkins controller, allowing attackers with Item/Configure permission to create or replace arbitrary files on the Jenkins controller file system with attacker-specified content, which can lead to remote code execution.

Jenkins Coverage Plugin 3.3358.v9487dde48783 and earlier does not validate the coverage results ID when a job configuration is submitted through the REST API, allowing attackers with Item/Configure permission to use a javascript: scheme URL as identifier, resulting in a stored cross-site scripting (XSS) vulnerability.

Jenkins OWASP Dependency-Check Plugin 5.6.4 and earlier does not escape CWE values from Dependency-Check reports on the Jenkins UI, resulting in a stored cross-site scripting (XSS) vulnerability exploitable by attackers with Item/Configure permission.

Jenkins Warnings Plugin 13.10258.va17d49a78c3b and earlier does not validate the analysis results ID when a job configuration is submitted through the REST API, allowing attackers with Item/Configure permission to use a javascript: scheme URL as identifier, resulting in a stored cross-site scripting (XSS) vulnerability.

Jenkins GitLab Plugin 1.2149.vcfc32c82bf7f and earlier caches the GitLab API client built for alternative GitLab API token credentials under a cache key derived from the credentials ID alone, omitting the folder in which the credentials are resolved, allowing attackers with Item/Configure permission to access GitLab API token credentials they are not entitled to use.

Jenkins Pipeline: Groovy Libraries Plugin 805.vafc79344957d and earlier does not restrict the library path provided to the library Pipeline step to a relative path inside the SCM checkout, and follows symbolic links to locations outside of the SCM checkout when retrieving the library, resulting in a path traversal vulnerability, allowing attackers able to configure Pipelines to read files in a resources directory and to delete files in a test directory on the Jenkins controller file system.

Jenkins Pipeline: Multibranch Plugin 841.vec5b9e1806ec and earlier does not set the appropriate context for credentials lookup in the resolveScm Pipeline step, allowing attackers with Item/Configure permission to access and capture credentials they are not entitled to.

Jenkins Script Security Plugin 1415.v9af9b3ac253d and earlier does not check calls from sandboxed scripts to methods added dynamically to a class at runtime, allowing attackers with permission to define and run sandboxed scripts, including Pipelines, to bypass the sandbox protection and execute code outside the sandbox.

Jenkins Script Security Plugin 1415.v9af9b3ac253d and earlier downloads a JAR file specified by URL twice, confirming the approval of the first download and loading the classpath entries from the second, allowing attackers able to define classpath entries to execute arbitrary code in the context of the Jenkins controller JVM.

Jenkins Script Security Plugin 1415.v9af9b3ac253d and earlier automatically approves the classpath entries in an item configuration when a user with Overall/Administer permission copies the item, or updates that configuration through the REST API or CLI, allowing attackers able to define classpath entries to execute arbitrary code in the context of the Jenkins controller JVM.

Jenkins Script Security Plugin 1415.v9af9b3ac253d and earlier does not reject the @GroovyASTTransformationClass annotation, allowing attackers with permission to define and run sandboxed scripts, including Pipelines, to run an arbitrary AST transformation at compile time, bypassing the sandbox protection and executing arbitrary code in the context of the Jenkins controller JVM.

Jenkins Script Security Plugin 1415.v9af9b3ac253d and earlier does not reject @Builder annotations whose builderStrategy member names an arbitrary class, allowing attackers with permission to define and run sandboxed scripts, including Pipelines, to execute code outside the sandbox if a suitable class is present on the classpath of the component that evaluates the script.