Latest Vulnerabilities with Public Exploits

Time-of-check time-of-use (toctou) race condition in Microsoft Windows QoS scheduler allows an authorized attacker to elevate privileges locally.

VMware ESXi and vCenter Server contain a reflected cross-site scripting vulnerability due to improper input validation. A malicious actor with network access to the login page of certain ESXi host or vCenter Server URL paths may exploit this issue to steal cookies or redirect to malicious websites.

A vulnerability in an API of Cisco ISE could allow an authenticated, remote attacker with valid read-only credentials to obtain sensitive information, change node configurations, and restart the node. This vulnerability is due to a lack of authorization in a specific API and improper validation of user-supplied data. An attacker could exploit this vulnerability by sending a crafted HTTP request to a specific API on the device. A successful exploit could allow the attacker to attacker to obtain information, modify system configuration, and reload the device. Note: To successfully exploit this vulnerability, the attacker must have valid read-only administrative credentials. In a single-node deployment, new devices will not be able to authenticate during the reload time.

imgproxy is server for resizing, processing, and converting images. Imgproxy does not block the 0.0.0.0 address, even with IMGPROXY_ALLOW_LOOPBACK_SOURCE_ADDRESSES set to false. This can expose services on the local host. This vulnerability is fixed in 3.27.2.

On April 18th 2025, Microsoft announced Exchange Server Security Changes for Hybrid Deployments and accompanying non-security Hot Fix. Microsoft made these changes in the general interest of improving the security of hybrid Exchange deployments. Following further investigation, Microsoft identified specific security implications tied to the guidance and configuration steps outlined in the April announcement. Microsoft is issuing CVE-2025-53786 to document a vulnerability that is addressed by taking the steps documented with the April 18th announcement. Microsoft strongly recommends reading the information, installing the April 2025 (or later) Hot Fix and implementing the changes in your Exchange Server and hybrid environment.

Teedy through 1.11 allows CSRF for account takeover via POST /api/user/admin.

A Server-Side Request Forgery (SSRF) vulnerability was identified in langgenius/dify version 0.10.2. The vulnerability occurs in the 'Create Knowledge' section when uploading DOCX files. If an external relationship exists in the DOCX file, the reltype value is requested as a URL using the 'requests' module instead of the 'ssrf_proxy', leading to an SSRF vulnerability. This issue was fixed in version 0.11.0.

Cursor is a code editor built for programming with AI. Cursor allows writing in-workspace files with no user approval in versions below 1.3.9, If the file is a dotfile, editing it requires approval but creating a new one doesn't. Hence, if sensitive MCP files, such as the .cursor/mcp.json file don't already exist in the workspace, an attacker can chain a indirect prompt injection vulnerability to hijack the context to write to the settings file and trigger RCE on the victim without user approval. This is fixed in version 1.3.9.

Electrolink FM/DAB/TV Transmitter Web Management System Unauthorized access vulnerability via the /FrameSetCore.html endpoint in Electrolink 500W, 1kW, 2kW Medium DAB Transmitter Web v01.09, v01.08, v01.07, and Display v1.4, v1.2.

The EG-Series plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the plugin's [series] shortcode in all versions up to, and including, 2.1.1 due to insufficient input sanitization and output escaping on user supplied attributes in the shortcode_title function. This makes it possible for authenticated attackers - with contributor-level access and above, on sites with the Classic Editor plugin activated - to inject arbitrary JavaScript code in the titletag attribute that will execute whenever a user access an injected page.

Claude Code is an agentic coding tool. In versions below 0.2.111, a path validation flaw using prefix matching instead of canonical path comparison, makes it possible to bypass directory restrictions and access files outside the CWD. Successful exploitation depends on the presence of (or ability to create) a directory with the same prefix as the CWD and the ability to add untrusted content into a Claude Code context window. This is fixed in version 0.2.111.

Improper neutralization of special elements used in an sql command ('sql injection') in Microsoft Configuration Manager allows an authorized attacker to execute code over an adjacent network.

File upload vulnerability in Writebot AI Content Generator SaaS React Template thru 4.0.0, allowing remote attackers to gain escalated privileges via a crafted POST request to the /file-upload endpoint.

GPMAW 14, a bioinformatics software, has a critical vulnerability related to insecure file permissions in its installation directory. The directory is accessible with full read, write, and execute permissions for all users, allowing unprivileged users to manipulate files within the directory, including executable files like GPMAW3.exe, Fragment.exe, and the uninstaller GPsetup64_17028.exe. An attacker with user-level access can exploit this misconfiguration by replacing or modifying the uninstaller (GPsetup64_17028.exe) with a malicious version. While the application itself runs in the user's context, the uninstaller is typically executed with administrative privileges when an administrator attempts to uninstall the software. By exploiting this flaw, an attacker could gain administrative privileges and execute arbitrary code in the context of the admin, resulting in privilege escalation.

A stored cross-site scripting (XSS) vulnerability in Anchor CMS v0.12.7 allows attackers to inject malicious JavaScript via the page description field in the page creation interface (/admin/pages/add).

Unisite CMS version 5.0 contains a stored Cross-Site Scripting (XSS) vulnerability in the "Report" functionality. A malicious script submitted by an attacker is rendered in the admin panel when viewed by an administrator. This allows attackers to hijack the admin session and, by leveraging the template editor, upload and execute a PHP web shell on the server, leading to full remote code execution.

A vulnerability was detected in givanz Vvveb 1.0.6.1. Impacted is an unknown function. The manipulation results in session fixiation. The attack can be launched remotely. The exploit is now public and may be used. Upgrading to version 1.0.7 is recommended to address this issue. The patch is identified as d4b1e030066417b77d15b4ac505eed5ae7bf2c5e. You should upgrade the affected component.

Cross site scripting vulnerability in seacms before 13.2 via the vid parameter to Upload/js/player/dmplayer/player.

/edit-user in webserver in OpenPLC Runtime 3 through 9cd8f1b allows authenticated users to upload arbitrary files (such as .html or .svg), and these are then publicly accessible under the /static URI.

The modelscope/ms-swift library thru 2.6.1 is vulnerable to arbitrary code execution through deserialization of untrusted data within the `load_model_meta()` function of the `ModelFileSystemCache()` class. Attackers can execute arbitrary code and commands by crafting a malicious serialized `.mdl` payload, exploiting the use of `pickle.load()` on data from potentially untrusted sources. This vulnerability allows for remote code execution (RCE) by deceiving victims into loading a seemingly harmless checkpoint during a normal training process, thereby enabling attackers to execute arbitrary code on the targeted machine. Note that the payload file is a hidden file, making it difficult for the victim to detect tampering. More importantly, during the model training process, after the `.mdl` file is loaded and executes arbitrary code, the normal training process remains unaffected'meaning the user remains unaware of the arbitrary code execution.