CVE Vulnerability Catalog

Cross Site Scripting (XSS) vulnerability in Fantastic Blog CMS 1.0. An attacker can inject a malicious script via the address field in pageEditMember.php.

A missing authorization check in GitLab CE/EE allows an unauthorized user to view project import source information. Affected versions are from 16.6 to 19.0.4, 19.1 to 19.1.2, and 19.2 to 19.2.0.

A vulnerability in linuxfabrik-lib before version 6.0.0 allows a redirecting server to intercept credential headers (other than Authorization and Cookie) during URL requests. The lib.url.fetch() function forwards headers on cross-origin redirects.

A vulnerability in linuxfabrik monitoring-plugins version 6.0.0 allows a local user controlling the plugin account to place a symlink that will be followed by sqlite3.connect() during a root-run log check. The database migration moves a file from /tmp using os.rename().

A vulnerability in MCP Ruby SDK before version 0.23.0 allows process memory exhaustion via unbounded creation of ServerSession objects in StreamableHTTPTransport. Sessions do not expire by default, and repeated initialize requests accumulate objects.

A vulnerability in MCP Ruby SDK before version 0.23.0 allows process memory exhaustion by sending data without a newline in StdioTransport and Stdio client. The IO#gets function is used without a byte limit.

A vulnerability in MCP Ruby SDK before version 0.23.0 allows DNS rebinding attacks due to missing validation of Host and Origin headers in StreamableHTTPTransport. A malicious web page can access a local MCP server and invoke tools.

A Directory Traversal vulnerability in DriveLock allows a remote, authenticated attacker to read sensitive files. The issue is in the web service listening on TCP port 4568, which does not properly validate user-supplied paths.

A Directory Traversal vulnerability in DriveLock allows a remote, unauthenticated attacker to read sensitive files. The issue is in the web service listening on TCP port 4568, which does not properly validate user-supplied paths.

GitLab CE/EE has an improper access control vulnerability. An authenticated user with guest role could, under certain conditions, access test report contents they were not authorized to view.

GitLab CE/EE has a Cross-Site Scripting (XSS) vulnerability. An attacker can execute arbitrary JavaScript in another user's browser via a crafted URL due to improper input sanitization.

An Open Redirect vulnerability in Dify AI Workflow's AppInitializer component. An attacker can redirect a user to a malicious site during OAuth flow, requiring user interaction (visiting a malicious page or file).

GitLab EE has a vulnerability in virtual registries where sensitive information could be disclosed to an unintended host due to improper handling of upstream requests.

GitLab EE has an authorization bypass vulnerability during token generation. An authenticated user could bypass administrator-configured tool governance policies.

GitLab EE has a vulnerability in the AI-assisted code review feature. An authenticated user could access information from unauthorized projects due to improper neutralization of untrusted content.

GitLab CE/EE has an authorization check vulnerability. An unauthenticated user could, under certain conditions, view the title of a confidential issue through a publicly accessible merge request.

GitLab CE/EE has an authorization vulnerability in a projects API endpoint. An authenticated user with Maintainer role could modify protected branch configuration due to improper authorization.

GitLab EE has a race condition in approval rule processing that under certain conditions allows an authenticated user to merge code into a protected branch without the required approvals. Affected versions are 17.0 through 19.0.4, 19.1 through 19.1.2, and 19.2 through 19.2.0.

pip incorrectly handles doubly-encoded package URLs from indexes, allowing files to be installed to arbitrary disk locations when installing wheels. This vulnerability requires downloading or installing a package from a malicious index; malicious packages alone cannot exploit it.

Netty prior to versions 4.1.136.Final and 4.2.16.Final does not escape or validate header values in STOMP CONNECT and CONNECTED frames, allowing an attacker to inject additional STOMP headers by controlling a header value (e.g., login or passcode). This can lead to authentication bypass or privilege escalation depending on the broker.