MotionEye video surveillance software version 0.43.1b4 and earlier contains an authenticated OS command injection via configuration parameters such as image_file_name. Admin users can inject commands that execute when the Motion daemon restarts, achieving code execution on the surveillance server.
UAF in Redis 8.2.1 via crafted Lua scripts by authenticated users. EPSS 12.4%. Patch available.
Authenticated Redis users with permission to run server-side Lua can trigger an integer overflow (CWE-190) in the bundled Lua interpreter by calling unpack() with crafted out-of-range indices, potentially escalating to remote code execution in Redis 8.2.1 and earlier. The issue affects all Redis versions that include Lua scripting, which is enabled by default, and carries a CVSS score of 8.8 (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H) reflecting high confidentiality, integrity, and availability impact but a low-privilege authenticated prerequisite. Publicly available exploit code exists, and EPSS is elevated at 13.20% (94th percentile), although active exploitation has not been confirmed by CISA KEV.
Authenticated Lua script execution in Redis 8.2.1 and earlier can be abused with a crafted long-string delimiter to trigger an out-of-bounds read or crash the server, yielding high confidentiality impact and denial of service. The flaw affects all Redis releases that support Lua scripting, which is enabled by default, but requires a credential whose ACL permits the EVAL and/or FUNCTION command families. Publicly available exploit code exists and the EPSS score is 5.05% (90th percentile), indicating elevated attention, but there is no confirmed active exploitation in CISA KEV; the authentication and scripting-permission barrier keeps this a moderate-priority issue rather than a critical one.
CVE-2025-46818 is a sandbox-escape flaw in Redis's embedded Lua scripting engine (all versions up to and including 8.2.1) that lets an authenticated user with EVAL/FUNCTION privileges poison shared Lua metatables for basic types such as strings and numbers, so that code they plant can execute later in another user's scripting context. Cross-user impact requires a second, more privileged user to run a Lua script against the same server instance after the tampering, and exploitation is fully blocked by denying the EVAL and FUNCTION command families via ACLs. Publicly available exploit code exists and EPSS sits at roughly 3.2% (87th percentile), but there is no confirmed active exploitation (not in CISA KEV); the practical exposure concentrates in multi-tenant or shared Redis deployments where multiple principals run scripts.
Auth bypass in RestroPress WordPress ordering plugin 3.0.0-3.1.9.2.
Duplicate-email registration handling in phpMyFAQ 4.0.7 through 4.0.13 (and 4.0-nightly-2025-10-03 and earlier) lets unauthenticated remote attackers create multiple distinct accounts sharing the same email address because the createUser logic in RegistrationHelper.php and User.php performed no uniqueness check prior to 4.0.13. The practical consequence is account ambiguity rather than immediate compromise: data-integrity loss, colliding password-reset flows, and spam/duplicate pollution, with privilege escalation or account takeover possible only in specific configurations where a privileged account already exists under the targeted email and downstream flows key off the email instead of the unique user ID. A public proof-of-concept and the upstream fix commit are available, EPSS is negligible (0.05%, 16th percentile), and there is no indication of confirmed active exploitation (no CISA KEV), although the NVD headline CVSS of 9.8 substantially overstates real-world risk relative to our assessment of AV:N/AC:H/PR:N/UI:N/C:L/I:L/A:N.
Eidos is an extensible framework for Personal Data Management. Versions 0.21.0 and below contain a one-click remote code execution vulnerability. An attacker can exploit this vulnerability by embedding a specially crafted eidos: URL on any website, including a malicious one they control. When a victim visits such a site or clicks on the link, the browser triggers the app’s custom URL handler (eidos:), causing the Eidos application to launch and process the URL, leading to remote code execution on the victim’s machine. This issue does not have a fix as of October 3, 2025
Code injection in Anthropic's Claude Code CLI before version 1.0.111 allows malicious code embedded in a project directory to execute before the startup trust dialog is accepted, giving an attacker arbitrary command execution with the privileges of the user who launched the tool (CVSS 4.0 base 8.7). Exploitation is unauthenticated (PR:N) but depends on the victim starting Claude Code inside an attacker-controlled, untrusted directory (UI:P in the 4.0 vector; UI:R under the assessed CVSS 3.1 vector, which scores scope-unchanged confidentiality, integrity and availability impact as High). Publicly available exploit code exists in the form of a public gist, but there is no evidence of confirmed active exploitation (not listed in CISA KEV) and EPSS is low at 0.05% (14th percentile), so real-world risk is concentrated among users who manually maintain Claude Code and remain below 1.0.111 rather than the auto-update population.
A security vulnerability in DirectAdmin (CVSS 8.2) that allows unauthorized attackers. Risk factors: public PoC available.
A TCL Smart TV running a vulnerable UPnP/DLNA MediaRenderer implementation is affected by a remote, unauthenticated Denial of Service (DoS) condition. By sending a flood of malformed or oversized SetAVTransportURI SOAP requests to the UPnP control endpoint, an attacker can cause the device to become unresponsive. This denial persists as long as the attack continues and affects all forms of TV operation. Manual user control and even reboots do not restore functionality unless the flood stops.
A flaw was found in QEMU.
CVE-2025-59489 is a security vulnerability (CVSS 7.4) that allows argument injection that can result. Risk factors: public PoC available.
Vertical privilege escalation in OpenSupports 4.11.0 allows a Level 1 staff member to edit supervised user lists of any account, enabling unauthorized access to ticket content of other users. The vulnerability has publicly available proof-of-concept exploit code, but no active exploitation in the wild has been reported. EPSS probability is low at 0.04%.
Server-side request forgery in OpenSupports 4.11.0 lets unauthenticated remote attackers force the backend to initiate arbitrary network connections to attacker-chosen destinations through two diagnostic endpoints that ship with permission => 'any'. Exploitation enables internal network scanning and interaction with services reachable from the OpenSupports host, but the assessed impact is limited to low confidentiality with no integrity or availability effect. Publicly available exploit code exists; EPSS is 0.05% (16th percentile) and there is no CISA KEV listing confirming active exploitation.
A stored Cross-Site Scripting (XSS) vulnerability has been discovered in MetInfo CMS version 8.0. The vulnerability exists due to insufficient validation and sanitization of SVG file uploads in the app\system\include\module\editor\Uploader.class.php component. This security flaw allows attackers to upload malicious SVG files containing JavaScript code that executes when the uploaded file is viewed or accessed.
nopCommerce 4.40.3 is vulnerable to XSS in the Product Name at /Admin/Product/Edit/[id]. Each time a user views the product in the shop, the XSS payload fires.
A stored Cross-Site Scripting (XSS) vulnerability has been discovered in MetInfo CMS version 8.0. The vulnerability exists in the download management module, specifically in the app\system\download\admin\download_admin.class.php component. The vulnerability allows attackers to upload malicious SVG files containing JavaScript code that executes when the uploaded file is viewed or accessed by users.
A stored Cross-Site Scripting (XSS) vulnerability has been discovered in MetInfo CMS version 8.0. The vulnerability exists due to insufficient validation and sanitization of SVG file uploads in the app\system\include\module\uploadify.class.php component, specifically in the website settings module. This security flaw allows attackers to upload malicious SVG files containing JavaScript code that executes when the uploaded file is viewed or accessed.
A stored Cross-Site Scripting (XSS) vulnerability has been discovered in Emlog Pro 2.5.19. The vulnerability exists due to insufficient validation of SVG file uploads in the /admin/media.php component, allowing attackers to upload malicious SVG files containing JavaScript code that executes when the uploaded file is viewed.
A stored Cross-Site Scripting (XSS) vulnerability has been discovered in XunRuiCMS version 4.7.1. The vulnerability exists due to insufficient validation of SVG file uploads in the dayrui/Fcms/Library/Upload.php component, allowing attackers to inject malicious JavaScript code that executes when the uploaded file is viewed.
A stored Cross-Site Scripting (XSS) vulnerability has been discovered in MetInfo CMS version 8.0. The vulnerability exists in the image management module, specifically in the app\system\img\admin\img_admin.class.php component. The vulnerability allows attackers to upload malicious SVG files containing JavaScript code that executes when the uploaded file is viewed or accessed by users.
A stored Cross-Site Scripting (XSS) vulnerability has been discovered in MetInfo CMS version 8.0. The vulnerability exists in the column management module, specifically in the app\system\column\admin\index.class.php component. The vulnerability allows attackers to upload malicious SVG files containing JavaScript code that executes when the uploaded file is viewed or accessed by users.
A stored Cross-Site Scripting (XSS) vulnerability has been discovered in Emlog Pro 2.5.19. The vulnerability exists in the email template configuration component located at /admin/setting.php?action=mail, which allows administrators to input HTML code that is not properly sanitized, leading to persistent JavaScript execution.
LFI in JoomSport WordPress plugin.
Auth bypass in Spirit Framework WordPress plugin <= 1.2.14. EPSS 0.46%.
Privilege escalation in Appy Pie Connect for WooCommerce via password reset.
Remote code execution in Cursor IDE 1.6.23 and below can be triggered by prompt injection that abuses case-insensitive path handling to modify protected files such as .cursor/mcp.json. The assessed vector is AV:N/AC:H/PR:N/UI:R/S:C/C:H/I:H/A:H because exploitation requires the Cursor AI agent to process attacker-controlled content and the host to use a case-insensitive filesystem, the default on Windows and macOS; Linux ext4/xfs hosts are generally not affected. EPSS is 0.14% (34th percentile), no public exploit was identified at time of analysis, and version 1.7 fixes the issue.
Uninitialized variable in Vigor Routers DrayOS HTTP CGI. EPSS 0.06%.
Stored cross-site scripting in Emlog 2.5.21 and earlier allows an authenticated user with access to the mail template settings to persist attacker-controlled JavaScript, which then executes when an administrator later opens the mail settings page, potentially enabling session or token theft and admin account takeover. Exploitation requires a valid backend account with mail-settings access plus administrator interaction; publicly available exploit code exists, but EPSS is low at 0.03% (9th percentile) and there is no CISA KEV confirmation of active exploitation. The vendor addressed the issue in Emlog 2.5.22.
Use-after-free in SVG pattern parsing — pattern node deleted but accessed later.
Stack overflow DoS in SVG rendering via recursive pattern elements.
SQL injection in Joomla mod_vvisit_counter v2.0.4j3.
SQL injection in WPRecovery WordPress plugin.
Stored cross-site scripting in Emlog Pro 2.5.21 and earlier allows an authenticated user who can post 'Twitter' messages to inject JavaScript that is saved server-side and runs in the browser of any user, including administrators, who clicks the malicious post. Publicly available exploit code exists, but EPSS is only 0.03% (7th percentile) and no CISA KEV entry is reported. No vendor fix is available, so risk depends on limiting posting privileges and avoiding untrusted posts.
An information disclosure vulnerability has been discovered in SeaCMS 13.1. The vulnerability exists in the admin_safe.php component located in the /btcoan/ directory. This security flaw allows authenticated administrators to scan and download not only the application’s source code but also potentially any file accessible on the server’s root directory.
The AP Background plugin for WordPress is vulnerable to arbitrary file uploads due to missing authorization and insufficient file validation within the advParallaxBackAdminSaveSlider() handler in versions 3.8.1 to 3.8.2. This makes it possible for authenticated attackers, with Subscriber-level access and above, to upload arbitrary files on the affected site's server which may make remote code execution possible.
Cursor is a code editor built for programming with AI. In versions 1.7 and below, automatic loading of project-specific CLI configuration from the current working directory (<project>/.cursor/cli.json) could override certain global configurations in Cursor CLI. This allowed users running the CLI inside a malicious repository to be vulnerable to Remote Code Execution through a combination of permissive configuration (allowing shell commands) and prompt injection delivered via project-specific Rules (<project>/.cursor/rules/rule.mdc) or other mechanisms. The fix for this issue is currently available as a patch 2025.09.17-25b418f. As of October 3, 2025 there is no release version.
Remote code execution in Cursor CLI Agent 1.7 and earlier can be triggered by unauthenticated attackers through prompt injection that overwrites the agent's trusted cli.json configuration on case-insensitive filesystems. Exploitation requires a developer to run the CLI agent against attacker-influenced content (CVSS UI:R), and the RCE path is gated on default case-insensitive filesystems such as macOS and Windows rather than typical Linux deployments. No public exploit identified at time of analysis, and EPSS is 0.12%, indicating low predicted exploitation likelihood.
Cursor is a code editor built for programming with AI. In versions 1.7 and below, when MCP uses OAuth authentication with an untrusted MCP server, an attacker can impersonate a malicious MCP server and return crafted, maliciously injected commands during the interaction process, leading to command injection and potential remote code execution. If chained with an untrusted MCP service via OAuth, this command injection vulnerability could allow arbitrary code execution on the host by the agent. This can then be used to directly compromise the system by executing malicious commands with full user privileges. This issue does not currently have a fixed release version, but there is a patch, 2025.09.17-25b418f.
An SQL injection vulnerability has been reported to affect Video Station. If a remote attacker gains a user account, they can then exploit the vulnerability to execute unauthorized code or commands. We have already fixed the vulnerability in the following version: Video Station 5.8.4 and later
The WP Dispatcher plugin for WordPress is vulnerable to SQL Injection via the ‘id’ parameter in all versions up to, and including, 1.2.0 due to insufficient escaping on the user supplied parameter and lack of sufficient preparation on the existing SQL query. This makes it possible for authenticated attackers, with Contributor-level access and above, to append additional SQL queries into already existing queries that can be used to extract sensitive information from the database.
The TextBuilder plugin for WordPress is vulnerable to Cross-Site Request Forgery in versions 1.0.0 to 1.1.1. This is due to missing or incorrect nonce validation on the 'handleToken' function. This makes it possible for unauthenticated attackers to update a user's authorization token via a forged request granted they can trick a site administrator into performing an action such as clicking on a link. Once the token is updated, an attacker can update the user's password and email address.
TCL 65C655 Smart TV, running firmware version V8-R75PT01-LF1V269.001116 (Android TV, Kernel 5.4.242+), is vulnerable to a blind, unauthenticated Server-Side Request Forgery (SSRF) vulnerability via the UPnP MediaRenderer service (AVTransport:1). The device accepts unauthenticated SetAVTransportURI SOAP requests over TCP/16398 and attempts to retrieve externally referenced URIs, including attacker-controlled payloads. The blind SSRF allows for sending requests on behalf of the TV, which can be leveraged to probe for other internal or external services accessible by the device (e.g., 127.0.0.1:16XXX, LAN services, or internet targets), potentially enabling additional exploit chains.
Karapace is an open-source implementation of Kafka REST and Schema Registry. Versions 5.0.0 and 5.0.1 contain an authentication bypass vulnerability when configured to use OAuth 2.0 Bearer Token authentication. If a request is sent without an Authorization header, the token validation logic is skipped entirely, allowing an unauthenticated user to read and write to Schema Registry endpoints that should otherwise be protected. This effectively renders the OAuth authentication mechanism ineffective. This issue is fixed in version 5.0.2.
SQL injection in QNAP Qsync Central 5.0.0 (all builds prior to 5.0.0.2) allows a remote attacker who already holds a valid, even low-privileged, user account to inject arbitrary SQL through the application's standard network interface, resulting in full compromise of the confidentiality and integrity of the backend database - potentially escalating to unauthorized code or command execution as the vendor notes. The vulnerability is not known to be actively exploited (no CISA KEV listing) and no public exploit code has been identified at time of analysis; EPSS is 0.09% (26th percentile), reflecting the mitigating requirement for valid credentials (CVSS v3.1 PR:L, AV:N, AC:L, UI:N). QNAP has released a fixed build, 5.0.0.2 (2025/07/31), which should be applied to eliminate the condition.
SQL injection in QNAP Qsync Central versions 5.0.0 through below 5.0.0.2 lets a remote attacker who already holds a valid Qsync Central user account run unauthorized code or commands on the underlying NAS with high confidentiality and integrity impact, though availability is not affected. The flaw is network-reachable and low-complexity but gated behind the account prerequisite (PR:L, authenticated), so unauthenticated internet exposure alone is not sufficient; EPSS is low (0.09%, 26th percentile) and no public exploit has been identified at time of analysis, with no CISA KEV listing. QNAP has released a vendor patch in Qsync Central 5.0.0.2 (2025/07/31).
NetBak Replicator 4.5.x on Windows allows an authenticated local user to execute arbitrary code via an unquoted search path, resulting in local privilege escalation. The flaw (CWE-428) affects versions before 4.5.15.0807 and requires an existing low-privileged account; execution occurs when the affected service or process starts or restarts, not instantaneously. No public exploit identified at time of analysis, and EPSS is low at 0.02%, though the vendor rates the issue High with a CVSS 4.0 score of 8.5.
DLL preloading in the Panasonic AutoDownloader 1.2.8 installer allows code execution when a crafted DLL is placed in the same directory and a user runs the installer from that location (CWE-427). The issue is strictly local: the attacker needs write access to the installer's execution directory and user interaction to launch it, so network reachability is absent and the installed/running product is not implicated. EPSS is 0.01% (2nd percentile); no CISA KEV entry or public exploit code is identified in the supplied intelligence, and the independent assessment rates real-world priority LOW despite the 8.4 CVSS 4.0 base score.
Improper Neutralization of Special Elements used in an SQL Command ('SQL Injection') vulnerability in Teknolojik Center Telecommunication Industry Trade Co.