Cybersecurity AI (CAI) is a framework for AI Security. In versions up to and including 0.5.10, the CAI (Cybersecurity AI) framework contains multiple argument injection vulnerabilities in its function tools. User-controlled input is passed directly to shell commands via `subprocess.Popen()` with `shell=True`, allowing attackers to execute arbitrary commands on the host system. The `find_file()` tool executes without requiring user approval because find is considered a "safe" pre-approved command. This means an attacker can achieve Remote Code Execution (RCE) by injecting malicious arguments (like -exec) into the args parameter, completely bypassing any human-in-the-loop safety mechanisms. Commit e22a1220f764e2d7cf9da6d6144926f53ca01cde contains a fix.

PsySH is a runtime developer console, interactive debugger, and REPL for PHP. Prior to versions 0.11.23 and 0.12.19, PsySH automatically loads and executes a `.psysh.php` file from the Current Working Directory (CWD) on startup. If an attacker can write to a directory that a victim later uses as their CWD when launching PsySH, the attacker can trigger arbitrary code execution in the victim's context. When the victim runs PsySH with elevated privileges (e.g., root), this results in local privilege escalation. This is a CWD configuration poisoning issue leading to arbitrary code execution in the victim user’s context. If a privileged user (e.g., root, a CI runner, or an ops/debug account) launches PsySH with CWD set to an attacker-writable directory containing a malicious `.psysh.php`, the attacker can execute commands with that privileged user’s permissions, resulting in local privilege escalation. Downstream consumers that embed PsySH inherit this risk. For example, Laravel Tinker (`php artisan tinker`) uses PsySH. If a privileged user runs Tinker while their shell is in an attacker-writable directory, the `.psysh.php` auto-load behavior can be abused in the same way to execute attacker-controlled code under the victim’s privileges. Versions 0.11.23 and 0.12.19 patch the issue.

Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection') vulnerability in TOTOLINK X6000R allows OS Command Injection.This issue affects X6000R: through V9.4.0cu.1498_B20250826.

# Active Storage allowed transformation methods potentially unsafe Active Storage attempts to prevent the use of potentially unsafe image transformation methods and parameters by default. The default allowed list contains three methods allow for the circumvention of the safe defaults which enables potential command injection vulnerabilities in cases where arbitrary user supplied input is accepted as valid transformation methods or parameters. Impact ------ This vulnerability impacts applications that use Active Storage with the image_processing processing gem in addition to mini_magick as the image processor. Vulnerable code will look something similar to this: ``` <%= image_tag blob.variant(params[:t] => params[:v]) %> ``` Where the transformation method or its arguments are untrusted arbitrary input. All users running an affected release should either upgrade or use one of the workarounds immediately. Workarounds ----------- Consuming user supplied input for image transformation methods or their parameters is unsupported behavior and should be considered dangerous. Strict validation of user supplied methods and parameters should be performed as well as having a strong [ImageMagick security policy](https://imagemagick.org/script/security-policy.php) deployed. Credits ------- Thank you [lio346](https://hackerone.com/lio346) for reporting this!

LobeHub is an open source human-and-AI-agent network. Prior to version 1.143.3, the file upload feature in `Knowledge Base > File Upload` does not validate the integrity of the upload request, allowing users to intercept and modify the request parameters. As a result, it is possible to create arbitrary files in abnormal or unintended paths. In addition, since `lobechat.com` relies on the size parameter from the request to calculate file usage, an attacker can manipulate this value to misrepresent the actual file size, such as uploading a `1 GB` file while reporting it as `10 MB`, or falsely declaring a `10 MB` file as a `1 GB` file. By manipulating the size value provided in the client upload request, it is possible to bypass the monthly upload quota enforced by the server and continuously upload files beyond the intended storage and traffic limits. This abuse can result in a discrepancy between actual resource consumption and billing calculations, causing direct financial impact to the service operator. Additionally, exhaustion of storage or related resources may lead to degraded service availability, including failed uploads, delayed content delivery, or temporary suspension of upload functionality for legitimate users. A single malicious user can also negatively affect other users or projects sharing the same subscription plan, effectively causing an indirect denial of service (DoS). Furthermore, excessive and unaccounted-for uploads can distort monitoring metrics and overload downstream systems such as backup processes, malware scanning, and media processing pipelines, ultimately undermining overall operational stability and service reliability. Version 1.143.3 contains a patch for the issue.

Improper Neutralization of Special Elements used in an Expression Language Statement ('Expression Language Injection') vulnerability in The Wikimedia Foundation Mediawiki - DiscussionTools Extension allows Regular Expression Exponential Blowup.This issue affects Mediawiki - DiscussionTools Extension: 1.44, 1.43.

SQL injection vulnerability in geopandas before v.1.1.2 allows an attacker to obtain sensitive information via the to_postgis()` function being used to write GeoDataFrames to a PostgreSQL database.

Salt contains an authentication protocol version downgrade weakness that can allow a malicious minion to bypass newer authentication/security features by using an older request payload format, enabling minion impersonation and circumventing protections introduced in response to prior issues.

Salt's junos execution module contained an unsafe YAML decode/load usage. A specially crafted YAML payload processed by the junos module could lead to unintended code execution under the context of the Salt process.

aelsantex runcommand 2014-04-01, a plugin for DokuWiki, allows unauthenticated attackers to execute arbitrary system commands via lib/plugins/runcommand/postaction.php.

Cleartext Storage of Sensitive Information vulnerability in OpenText™ Vertica allows Retrieve Embedded Sensitive Data.   The vulnerability could read Vertica agent plaintext apikey.This issue affects Vertica versions: 23.X, 24.X, 25.X.

A vulnerability was detected in SourceCodester Pet Grooming Management Software 1.0. Impacted is an unknown function of the file /admin/operation/user.php of the component User Management. Performing a manipulation of the argument group_id results in improper authorization. The attack can be initiated remotely. The exploit is now public and may be used.

A security vulnerability has been detected in itsourcecode Student Management System 1.0. This issue affects some unknown processing of the file /enrollment/index.php. Such manipulation of the argument ID leads to sql injection. It is possible to launch the attack remotely. The exploit has been disclosed publicly and may be used.

Insufficient epoch key slot processing in OpenVPN 2.7_alpha1 through 2.7_rc5 allows remote authenticated users to trigger an assert resulting in a denial of service

A weakness has been identified in projectworlds House Rental and Property Listing 1.0. This vulnerability affects unknown code of the file /app/sms.php. This manipulation of the argument Message causes cross site scripting. It is possible to initiate the attack remotely. The exploit has been made available to the public and could be used for attacks.

A vulnerability has been found in bolo-solo up to 2.6.4. This impacts the function importMarkdownsSync of the file src/main/java/org/b3log/solo/bolo/prop/BackupService.java of the component SnakeYAML. Such manipulation leads to deserialization. The attack may be launched remotely. The exploit has been disclosed to the public and may be used.

A flaw has been found in Tenda HG10 US_HG7_HG9_HG10re_300001138_en_xpon. This affects the function system of the file /boaform/formSysCmd. This manipulation of the argument sysCmd causes command injection. The attack may be initiated remotely. The exploit has been published and may be used.

A vulnerability was detected in Tenda HG10 US_HG7_HG9_HG10re_300001138_en_xpon. The impacted element is the function checkUserFromLanOrWan of the file /boaform/admin/formLogin of the component Login Interface. The manipulation of the argument Host results in command injection. The attack can be launched remotely. The exploit is now public and may be used.

Atomic Alarm Clock 6.3 contains a local privilege escalation vulnerability in its service configuration that allows attackers to execute arbitrary code with SYSTEM privileges. Attackers can exploit the unquoted service path by placing a malicious executable named 'Program.exe' to gain persistent system-level access.

Popcorn Time 6.2.1.14 contains an unquoted service path vulnerability that allows local non-privileged users to potentially execute code with elevated system privileges. Attackers can insert malicious executables in Program Files (x86) or system root directories to be executed with SYSTEM-level permissions during service startup.

Andrea ST Filters Service 1.0.64.7 contains an unquoted service path vulnerability in its Windows service configuration. Local attackers can exploit the unquoted path to inject malicious code that will execute with elevated LocalSystem privileges during service startup.

Outline Service 1.3.3 contains an unquoted service path vulnerability that allows local users to potentially execute arbitrary code with elevated system privileges. Attackers can exploit the unquoted binary path in C:\Program Files (x86)\Outline to inject malicious code that would execute with LocalSystem permissions during service startup.

OpenZ ERP 3.6.60 contains a persistent cross-site scripting vulnerability in the Employee module's name and description parameters. Attackers can inject malicious scripts through POST requests to , enabling session hijacking and manipulation of application modules.

Orchard Core RC1 contains a persistent cross-site scripting vulnerability that allows remote attackers to inject malicious scripts through blog post creation. Attackers can create blog posts with embedded JavaScript in the MarkdownBodyPart.Source parameter to execute arbitrary scripts in victim browsers.

Tryton 5.4 contains a persistent cross-site scripting vulnerability in the user profile name input that allows remote attackers to inject malicious scripts. Attackers can exploit the vulnerability by inserting script payloads in the name field, which execute in the frontend and backend user interfaces.