CVE Vulnerability Catalog

A security vulnerability was detected in JumpServer up to 4.10.19 in an unknown function of the Template Rendering component; manipulation of the IP/HOST argument leads to OS command injection. Remote exploitation is possible, the exploit has been publicly disclosed, and upgrading to 5.0.0 fixes the issue (patch fcf2904cdb884230e9676f50f37a44d09fa97040); the vendor confirms the fix will also be included in an upcoming v4 release.

In aiohttp before 3.14.4, the pure Python parser accepts ambiguous HTTP request framing. An unauthenticated attacker can smuggle requests when an application runs without a standard wheel (e.g., with AIOHTTP_NO_EXTENSIONS) and an intermediary interprets the framing differently.

In aiohttp before 3.14.4, an attacker can send an absolute-form request causing Request.scheme to use attacker-controlled request-target scheme data instead of the connection protocol. Applications using Request.scheme for security decisions can be tricked about whether the connection is encrypted.

In aiohttp before 3.14.4, Max-Age processing can overflow and skip closing a connection. An attacker can temporarily exhaust the connection pool until garbage collection resolves the connections.

In aiohttp before 3.14.4, the Secure attribute for a shared cookie is not enforced. An attacker can cause a shared cookie marked Secure to be sent with an unencrypted request.

In russh before 0.64.1, the agent server spawns a separate long-lived task for every SSH_AGENT_CONSTRAIN_LIFETIME record in a constrained add-key request. The loop accepts repeated records, clones the key blob into each task, and does not roll back tasks when later constraint parsing fails, allowing an attacker to exhaust memory and tasks.

In Mercury from 3.0.0 until 3.2.11, mr.Markdown converts attacker-controlled Markdown to HTML without sanitizing raw HTML before assigning it to an ipywidgets.HTML sink. Event-handler attributes and other active HTML can reach the DOM and execute when another user views the application.

The Management Console and product files expose sensitive data, including remote log server credentials, truststore passwords, and keystore passwords, in plain text. This occurs because the system does not sufficiently protect these credentials when the remote logging feature is enabled.

The ForgeRock client fails to correctly validate refresh tokens. A refresh token can be treated as a valid access token, enabling its use for consuming protected resources and bypassing the requirement for valid client credentials.

The carbon console accepts user-supplied XML input without sufficiently encoding potentially executable content, allowing injection of malicious scripts. Successful exploitation allows a remote attacker to execute arbitrary JavaScript in the victim's browser context.

The authentication endpoint of WSO2 Identity Server fails to provide a consistent system response when a username does not exist, allowing attackers to differentiate between valid and invalid usernames. This difference can be leveraged to enumerate existing user accounts.

Tokens associated with user sessions are not revoked when the session is terminated, unless another SSO request is initiated. This occurs when session-bound token validation is enabled and the "Revoke token upon user logout" option is inactive.

A gap in the Netskope Client (NSClient) allows a system connecting to a Captive Portal access point to switch to a network with internet connectivity without closing the Captive Portal. This enables abuse via a rogue access point to access malicious sites or exfiltrate data.

Crawl4AI from 0.8.7 through 0.9.5 contains a server-side request forgery vulnerability in the Docker API server that allows unauthenticated attackers to bypass destination checks via URL userinfo parsing differences. Attackers can set fetch_ssl_certificate=true with crafted userinfo URLs so SSLCertificate.from_url connects to internal hosts on port 443, probing reachability and leaking certificate details.

docling-core 2.5.0 before 2.103.0 contains a local file disclosure vulnerability that allows attackers to read local images by supplying plain filesystem paths in ImageRef URIs, bypassing the allow_image_file_uri gate. Attackers can submit crafted DoclingDocument JSON or YAML that, when exported with ImageRefMode.EMBEDDED or rendered to HTML, embeds readable local image files as base64 and reveals path existence.

GDevelop desktop app through 5.6.284 contains a credential leak vulnerability that allows attackers to steal the gd_resource cookie by supplying attacker-controlled resource URLs in project files. Attackers can get victims to open a crafted project, causing downloadLocalFile() to send the cookie to their server and expose private cloud project resources.

A weakness has been identified in MikroTik RouterOS up to 6.40.5/6.44/7.20.8. The same_check function in ipt_SAME.ko can lead to an integer overflow. The attack may be launched remotely.

FOSSBilling from 0.1.0 until 0.8.7 has an unauthenticated POST /api/guest/client/create endpoint that reveals whether a submitted email belongs to a registered client by returning a distinct unredacted error. An attacker can compare responses to determine if an email belongs to an existing client.

FOSSBilling from 0.1.0 until 0.8.7 has a path traversal vulnerability in the unauthenticated api/guest/system/template_exists endpoint because templateExists uses Path::join on the attacker-controlled file parameter without rejecting .. sequences or verifying the resolved path stays inside the active theme directory. An attacker can determine whether arbitrary files or directories exist on the server.

Camunda 7.24.0 before 7.24.16 and 7.23.0 before 7.23.17 contains an authorization bypass vulnerability in the Cockpit webapp that allows authenticated attackers to bypass permission checks due to an inverted comparison logic in AbstractCockpitPluginResource.isPermissionDisabled().