CVE Feed

A flaw has been found in agentscope-ai AgentScope up to 2.0.8. This affects the function check_permissions of the file src/agentscope/tool/_builtin/_bash.py of the component Bash Tool. This manipulation causes incorrect authorization. Remote exploitation of the attack is possible. The exploit has been published and may be used. Upgrading to version 2.0.9 mitigates this issue. Patch name: 6b43b13a46401e86abd3defeed9a8bfe5b37bc6d. It is recommended to upgrade the affected component.

A flaw was found in the SAML HTTP-Redirect binding implementation of Keycloak. The issue occurs because the server processes percent-encoded parameter names (like %53AMLResponse) while the signature verification logic only checks literal parameter names. An attacker can use this discrepancy to provide a crafted, unsigned SAML message that the server processes instead of a legitimately signed one, leading to full account takeover of federated users.

A flaw was found in the management interface of the Quarkus framework, specifically within the quarkus-vertx-http component. The issue occurs when the framework incorrectly combines the management root path with specific permission rules, leading to a situation where security policies are applied to the wrong internal paths. As a result, an attacker can bypass intended authorization checks and gain unauthorized access to sensitive management endpoints.

A vulnerability was detected in agentscope-ai AgentScope up to 2.0.11. The impacted element is the function subprocess.Popen of the file src/agentscope/mcp/_config.py of the component MCP Workspace API. The manipulation of the argument command results in os command injection. The attack may be launched remotely. The exploit is now public and may be used. The pull request to fix this issue awaits acceptance.

Improper neutralization of special elements used in an SQL command ('SQL injection') vulnerability in Kobikom Telecommunications and Informatics Services Inc. Kobikom allows SQL Injection. This issue affects Kobikom: before KobiPBX-FOP3-SEC-r1.

In Eclipse BaSyx Helm Charts version 2.2.0 of the `aas-basyx-v2-full` chart, deployments with the bundled Keycloak enabled store the client secret of the `basyx-technical-user` OAuth2 client in cleartext in a Kubernetes ConfigMap. This client uses the client-credentials grant and its service account holds the realm `admin` role. The chart correctly writes `keycloak.technicalUser.clientSecret` to a Kubernetes Secret, but the template `basyx-keycloak-realm.yaml` also substitutes the same value into the Keycloak realm-import JSON, which is stored in the ConfigMap `<keycloak.fullnameOverride>-realm`. A principal that can read ConfigMaps but not Secrets in the release namespace, such as a holder of the built-in Kubernetes `view` ClusterRole, can read the client secret from the ConfigMap. They can then exchange it at the Keycloak token endpoint for an access token with the `admin` role. That token grants full create, read, update, delete and execute rights on all BaSyx repositories and registries in the deployment. The bundled Keycloak is disabled by default. No fixed release of the `aas-basyx-v2-full` chart exists. Affected users should migrate to the `basyx` chart, or deploy without the bundled Keycloak and rotate the `basyx-technical-user` client secret.

Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal') vulnerability in Kobikom Telecommunications and Informatics Services Inc. Kobikom allows Path Traversal. This issue affects Kobikom: before KobiPBX-FOP3-SEC-r1.

The Symfony HttpClient component NoPrivateNetworkHttpClient does not pin the checked IP address of cached hosts, allowing for DNS rebinding to bypass private network restrictions, resulting in enumeration of internal hosts and ports in Symfony versions prior to 5.4.56, 6.x prior to 6.4.50, 7.x prior to 7.4.23, and 8.x prior to 8.1.11. NoPrivateNetworkHttpClient resolves the host of each request and checks the IP before connecting. It caches the result per host, and passes the IP to the decorated client through the resolve option so that the client connects to the exact IP that was checked. The issue is that resolve was only passed on the first lookup of a host: on subsequent requests, the decorated client resolved the host again on its own. An attacker who controls a host's DNS can exploit this with DNS rebinding by answering the first lookup with a public IP, and subsequent lookups with a private IP, so when an application sends several requests to that host through the same client, the later requests connect to the private IP. This happens, for example, when an application fetches several URLs of a repository provided by a user. The check of the primary_ip in the on_progress callback still aborts the request before anything is sent, but the connection is made, and the error differs depending on whether the port is open or closed. This allows enumerating internal hosts and ports, as in CVE-2024-50342.

A security vulnerability has been detected in eosphoros-ai DB-GPT up to 0.8.1. The affected element is the function FileBasedSkill.get_prompt of the file packages/dbgpt-core/src/dbgpt/agent/skill/claude_skill/__init__.py of the component Jinja2 Template Handler. The manipulation leads to improper neutralization of special elements used in a template engine. The attack may be initiated remotely. The exploit has been disclosed publicly and may be used. Upgrading to version 0.8.2 is sufficient to fix this issue. The identifier of the patch is a9e7a3c026d114fcfb4ffe3eaf78851a636ded16. It is suggested to upgrade the affected component.

Inconsistent interpretation of HTTP requests ('HTTP Request/Response smuggling') vulnerability in Apache Sling Engine. The engine's query/form parser accepts non-standard percent-escapes, wrong charsets, and continues request processing on such errors. This issue affects Apache Sling Engine: before 3.0.4. Users are recommended to upgrade to version 3.0.4, which fixes the issue.

Improper output neutralization for logs in Apache Sling Engine. This issue affects Apache Sling Engine: before 3.0.4. Users are recommended to upgrade to version 3.0.4, which fixes the issue.

Improper neutralization of input during web page generation ('cross-site scripting') vulnerability in Apache Sling Engine. Content-Type, charset, and redirect changes from an included component evade the  protectheaders / checkcontenttype  enforcement through setHeader(), repeated calls to setContentType(), the charset APIs, Servlet 6.1 redirect overloads, and stale config - enabling stored XSS despite enabled protections. This issue affects Apache Sling Engine before 3.0.4. Users are recommended to upgrade to version 3.0.4, which fixes the issue.

Execution with unnecessary privileges vulnerability in TÜBİTAK BİLGEM Software Technologies Research Institute pardus-usb-formatter allows Privilege Abuse. This issue affects pardus-usb-formatter: from 0.7.0 before 0.7.2.

In Eclipse BaSyx Helm Charts versions 3.0.0-rc.1 to 3.12.1 of the `basyx` chart, deployments with the bundled Keycloak enabled store the passwords of the Keycloak users that the chart creates in cleartext in a Kubernetes ConfigMap. The template `keycloak-init-config.yaml` rendered `keycloak.initialization.users`, including each user's `credentials[].value`, as `users.json` into the persistent ConfigMap `<release>-basyx-keycloak-init-config`. By default this includes the bootstrap administrator user. That user holds the `admin` role and has a non-temporary password. A principal that can read ConfigMaps but not Secrets in the release namespace, such as a holder of the built-in Kubernetes `view` ClusterRole, can read the password from the ConfigMap and log in to Keycloak as this user. The resulting token carries the `admin` role, which the default BaSyx ABAC rules map to full rights on all BaSyx APIs in the deployment. The bundled Keycloak is disabled by default. The issue is fixed in version 3.12.2 of the `basyx` chart, which moves the initialization users to the Keycloak credentials Secret. Affected users should upgrade and change the passwords of users that were created through the chart.

A weakness has been identified in eosphoros-ai DB-GPT up to 0.8.2. Impacted is the function exec_module of the file hf_adapter.py of the component HuggingFace Adapter. Executing a manipulation of the argument trust_remote_code can lead to inclusion of functionality from untrusted control sphere. The attack can be launched remotely. The exploit has been made available to the public and could be used for attacks. The project was informed of the problem early through an issue report but has not responded yet.

Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal') vulnerability in Apache Juneau. The microservice DirectoryResource does not keep request paths inside its configured root directory. When an application enables the resource's view, upload or delete settings (all off by default), a client can read, write or delete files outside the root, anywhere the server process has access. This issue affects Apache Juneau: through 9.2.0. Users are recommended to upgrade to version 9.2.1, which fixes the issue.

Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') vulnerability in Apache Juneau. HTML document serialization does not HTML-encode variable substitutions in the page chrome (header, nav, aside, footer and similar fields). Applications whose chrome includes request-derived variables, such as request headers, form data or path values, reflect those values unencoded, allowing cross-site scripting. The default chrome does not use request variables, so only applications that add them are affected. This issue affects Apache Juneau: through 9.2.0. Users are recommended to upgrade to version 9.2.1, which fixes the issue.

Improper Neutralization of CRLF Sequences in HTTP Headers ('HTTP Response Splitting') vulnerability in Apache Juneau. The Apache Juneau REST server passed response header values and redirect targets to the servlet container unchecked. An application that sets a response header or redirect target from request data could let a remote client inject header lines (where the servlet container does not itself reject CR/LF) or redirect users to an arbitrary external site. The framework itself does not build headers or redirects from request data; exposure depends on application code. This issue affects Apache Juneau: through 9.2.0. Users are recommended to upgrade to version 9.2.1, which fixes the issue. Header values are now sanitized by default; applications that build redirect targets from request data must also switch from sendRedirect() to the new sendSafeRedirect().

Insufficiently Protected Credentials vulnerability in Apache Juneau. The Apache Juneau RestClient followed HTTP redirects while keeping credential headers from the original request, so a server that redirected the client to a different host or port, or from HTTPS to HTTP, could receive headers such as Authorization, Cookie, Proxy-Authorization and X-API-Key. This issue affects Apache Juneau: through 9.2.0. Users are recommended to upgrade to version 9.2.1, which fixes the issue. The fix applies to HTTP clients created by the RestClient builder; applications that supply their own HttpClient must disable automatic redirects or strip credential headers on cross-origin redirects themselves.

Generation of Error Message Containing Sensitive Information vulnerability in Apache Juneau. The REST server sends the internal exception class and message to clients in a "Thrown" response header on every error response, and can include internal exception messages in the error body, even when stack-trace rendering is disabled. This reveals implementation details to remote clients. This issue affects Apache Juneau: through 9.2.0. Users are recommended to upgrade to version 9.2.1, which fixes the issue.

Insertion of Sensitive Information into Log File vulnerability in Apache Juneau. The default REST server call logger records request headers for requests that end in a server error, without redacting them. Any client able to cause a server error gets headers such as Authorization, Cookie, Proxy-Authorization and X-API-Key written in cleartext to the application log, exposing them to anyone who can read the logs. This issue affects Apache Juneau: through 9.2.0. Users are recommended to upgrade to version 9.2.1, which fixes the issue.

Cross-Site Request Forgery (CSRF) vulnerability in Apache Juneau. The REST server accepts a "content" query parameter that replaces the request body, and this is enabled by default on every HTTP method, including GET. A cross-site link can therefore deliver an attacker-chosen request body to an endpoint without a CORS preflight, defeating protections that rely on the HTTP method or content type instead of an anti-CSRF token. This issue affects Apache Juneau: through 9.2.0. Users are recommended to upgrade to version 9.2.1, which fixes the issue. Applications that do not use the content parameter can also set disableContentParam to true.

Improper control of dynamically-loaded code vulnerability in Apache Juneau. When Apache Juneau parses a value into a property of type java.lang.Class (including parameterized forms), the default class swap loaded the named class with initialization enabled, so input naming a class on the application classpath ran that class's static initializer. The impact depends on the classes present in the deployment. This issue affects Apache Juneau: through 9.2.0. Users are recommended to upgrade to version 9.2.1, which fixes the issue.

Allocation of Resources Without Limits or Throttling vulnerability in Apache Juneau. The MessagePack parser, which the REST server enables by default, sizes buffers from length values in the input. An unauthenticated client can send a few bytes that force a multi-gigabyte allocation and an OutOfMemoryError, causing denial of service. Related unbounded allocations in REST request handling (multipart uploads and request reader buffers) are fixed at the same time. This issue affects Apache Juneau: through 9.2.0. Users are recommended to upgrade to version 9.2.1, which fixes the issue.

A security flaw has been discovered in eosphoros-ai DB-GPT up to 0.8.2. This issue affects the function execute_code of the file code_utils.py of the component CodeAction. Performing a manipulation results in code injection. The attack can be initiated remotely. The exploit has been released to the public and may be used for attacks. The project was informed of the problem early through an issue report but has not responded yet.

A flaw was found in ansible-core. When a task marked `no_log: true` fails, the task result is censored before being emitted, but the `exception` field is retained because it is present in the PRESERVE allow-list in lib/ansible/_internal/_task.py, and the default stdout callback prints it through CallbackBase._handle_exception. As a result, sensitive values the operator explicitly asked to hide can be disclosed: module argument-validation errors echo the argument value, templating errors print an Origin source snippet (the failing line plus the two preceding lines of the variables file, exposing neighbouring secrets from an -e @secrets.yml file), and fail/assert messages are printed in full. Secret material supplied as extra variables or variables files can therefore leak to terminal output, CI logs, and the stdout logs of any automation wrapping ansible-playbook, even though every task touching the secrets is marked no_log: true. This is a regression introduced with the ansible-core 2.19 data-tagging/templating rework; the behaviour does not occur on 2.16, where the error text is correctly censored.

A flaw was found in the ansible-core user module. When generate_ssh_key is enabled, User.ssh_key_gen() guards the key files against symlinks but checks the ~/.ssh directory only with os.path.exists(). An unprivileged user on a managed node who replaces their ~/.ssh with a symlink to an existing root-writable directory causes the privileged module to write the generated key pair through the symlink and then chown both files to that user. Because such files can be dropped into directories processed by schedulers (e.g. /etc/cron.daily) or dynamic MOTD (/etc/update-motd.d), the user can escalate to root on the managed node. This is an incomplete fix of CVE-2024-9902.

An OS command injection vulnerability exists in SHIRASAGI, which may allow an authenticated attacker to execute arbitrary OS commands on the system running the product.

A cross-site scripting vulnerability exists in SHIRASAGI, which may allow an authenticated attacker to execute arbitrary script in a user's web browser when the user accesses a website using the affected product.

vault-java-driver, a Java client for HashiCorp Vault, parses Vault's JSON responses with a vendored minimal-json parser that has no nesting depth limit. A deeply nested value raises a StackOverflowError, which escapes catch (Exception) handlers and kills the reading thread. A Vault token that can write but not read a KV v2 path can store such a value there, so every later read of that path through the driver fails. This requires Vault 1.19.x or earlier, because 1.20 and later reject values nested deeper than 298 levels. This issue affects io.github.jopenlibs:vault-java-driver 5.2.0 through 6.2.3 and is fixed in 7.0.0.

vault-java-driver, a Java client for HashiCorp Vault, silently disables TLS certificate and hostname verification in three states where the caller asked for verification: an SslConfig passed to VaultConfig without calling build(), an SslConfig restored by Java deserialization, and VAULT_SSL_VERIFY set to a truthy value other than "true", such as 1, yes or on. In each case SslConfig.isVerify() returns false and the driver installs a trust-all SSLContext. An attacker on the network path to the Vault server can then read the Vault token and secrets in transit and forge responses. This issue affects io.github.jopenlibs:vault-java-driver 5.2.0 through 6.2.3. It is fixed in 7.0.0.

The Dokan WordPress plugin before 5.3.3 does not reconcile the amount credited to a vendor's reverse-withdrawal ledger against the amount actually collected for the payment order, allowing a vendor to clear their outstanding commission debt while paying less than the balance owed when a discount coupon is applied at checkout.

The Security For WooCommerce WordPress plugin from 1.0.10 to 1.0.16 does not perform capability checks when creating users through an action available to unauthenticated visitors, and lets the caller choose the new account's role, allowing unauthenticated attackers to create administrator accounts and take over the site.

Authentication bypass in the HA heartbeat origin check in middlewared in TrueNAS SCALE before 25.10.8 (and in release trains before their respective fixes) allows an attacker on the same network segment to achieve unauthenticated full-administrator API access and execute commands as root. middlewared registers the HA peer-authentication hook (ha_permission) unconditionally, with no failover-licence or heartbeat-interface check. An unauthenticated connection whose TCP peer address is a hardcoded HA heartbeat address (169.254.10.1 or 169.254.10.2) and whose peer port is <= 1024 is granted a node credential that is authorized for every API method and treated as full administrator. Both conditions are properties of the attacker's own socket, and thus an attacker on the same layer-2 segment satisfies them without forging anything. The hook is present on non-HA, unlicensed, single-node Community Edition installations. The fix additionally requires that the connection terminate on the HA heartbeat listener (a local address/port the attacker cannot control).

Wowza Streaming Engine before 4.12.2 processes XML external entities, which allows remote attackers to read local files with the privileges of the Wowza Streaming Engine process.

An OS Command Injection vulnerability was discovered in the Google Cloud DNS validation module of the OPNsense os-acme-client plugin through 4.17. The DnsGcloud::prepare() method decodes the dns_gcloud_key model field and directly interpolates the string fields gcloud_account and gcloud_project into a shell command passed to proc_open() via LeUtils::run_shell_command() without any validation or escaping. A remote authenticated user with privileges to issue or renew certificates through the WebGUI/API can achieve arbitrary code execution as root.

GodotSteam 4.17 through 4.23.2 trusts the peer value of 1 to mean that a player is the server. A grant_gold(1000000) call can be forged.

The WP PageNavi Style WordPress plugin through 1.4 does not sanitise and escape some of its settings, and has no CSRF check when they are saved, which could allow attackers to make a logged-in administrator store arbitrary JavaScript via a CSRF attack that is then executed in the browser of everyone who visits the site's front end.

The CatFolders WordPress plugin before 2.5.7 does not have authorisation checks on a folder maintenance action, allowing users with a role as low as subscriber to reset the per-user ownership of every media folder and disable the CatFolders WordPress plugin before 2.5.7's user restriction setting.

The CatFolders WordPress plugin before 2.5.7 does not properly sanitize and escape a user-controlled sort setting before using it in a SQL query, allowing authenticated users with upload capability (Author and above) to perform SQL injection attacks.

The Dokan: AI Powered WooCommerce Multivendor Marketplace Solution WordPress plugin before 5.3.3 does not verify that the acting administrator is authorized to manage each selected user before enabling that user's marketplace seller status, allowing a subsite Administrator on WordPress Multisite to enable selling for users outside their own site, including network-level users they cannot otherwise manage.

The Dokan: AI Powered WooCommerce Multivendor Marketplace Solution WordPress plugin before 5.3.3 does not use a trusted site-owned sender address when relaying its store contact form, instead placing the unauthenticated submitter's supplied address directly in the message's From header rather than a Reply-To, allowing unauthenticated attackers to make the site email its own vendors with the sender spoofed to an address at the site's own domain (such as its administrator), aligned with the domain's own SPF/DKIM.

The Dokan: AI Powered WooCommerce Multivendor Marketplace Solution WordPress plugin before 5.3.3 does not correctly verify authorization when serving a single order and its order notes, allowing unauthenticated attackers to read customer and order information and to add or delete order notes for a subset of orders.

The Car Dealer (Dealership) and Vehicle sales WordPress plugin through 4.73 does not restrict its unauthenticated single-listing AJAX action to published content, allowing an unauthenticated attacker to read the title, content and metadata of any private, draft, pending, future or trashed post by supplying its numeric id.

The Easy Social Icons WordPress plugin before 4.0.3 does not sanitise SVG files, which it adds to the list of file types that may be uploaded site-wide, allowing users with the Author role and above to upload a malicious SVG containing XSS payloads. Where another installed Easy Social Icons WordPress plugin before 4.0.3 exposes an upload route to unauthenticated visitors, the same issue can be triggered without an account.

The IVPAY payments WordPress plugin through 1.0.8 does not sanitise a value from its payment callback before using it in a SQL statement, and treats an empty API key, which is its own default, as valid authentication, allowing unauthenticated attackers to read arbitrary data from the database.

The Cloud Blocks WordPress plugin through 1.1.8 does not perform any authorisation or sanitisation when storing block asset URLs, allowing unauthenticated users to store malicious scripts that are executed in the browser of any privileged user, such as an administrator, who opens the block editor.

The Easy Social Icons WordPress plugin before 4.0.3 does not sanitise and escape some of its settings before outputting them on front-end pages, which could allow high-privilege users such as Administrators to perform Stored Cross-Site Scripting attacks when the unfiltered_html capability is disallowed, for example on a multisite network.

The EWWW Image Optimizer WordPress plugin before 8.9.1 does not properly sanitize and restrict a value in post content before using it to dynamically load and execute a script in the page, allowing authenticated attackers with author-level access and above to inject arbitrary web scripts that run in the session of any user who views the affected page, including administrators. This is an incomplete fix of CVE-2026-15446: the origin and path restrictions added from 8.7.4 onward were each bypassable, so versions 8.7.4 through 8.9.0 remain affected.

The Essential Real Estate WordPress plugin before 5.3.5 does not verify ownership of an invoice before rendering it in one of its AJAX actions, allowing an authenticated low-privileged user to read other users' invoices, including the invoice owner's name, account email address and financial details.