Langroid is a Python framework to build large language model (LLM)-powered applications. Prior to version 0.53.15, `TableChatAgent` uses `pandas eval()`. If fed by untrusted user input, like the case of a public-facing LLM application, it may be vulnerable to code injection. Langroid 0.53.15 sanitizes input to `TableChatAgent` by default to tackle the most common attack vectors, and added several warnings about the risky behavior in the project documentation.

Langroid is a framework for building large-language-model-powered applications. Prior to version 0.59.32, there is a bypass to the fix for CVE-2025-46724. TableChatAgent can call pandas_eval tool to evaluate the expression. There is a WAF in langroid/utils/pandas_utils.py introduced to block code injection CVE-2025-46724. However it can be bypassed due to _literal_ok() returning False instead of raising UnsafeCommandError on invalid input, combined with unrestricted access to dangerous dunder attributes (__init__, __globals__, __builtins__). This allows chaining whitelisted DataFrame methods to leak the eval builtin and execute arbitrary code. This issue has been patched in version 0.59.32.

A CWE-94: Improper Control of Generation of Code ('Code Injection') vulnerability exists that allows remote code execution via the “hostname” parameter when maliciously crafted hostname syntax is entered. Affected products: StruxureWare Data Center Expert (V7.9.2 and prior)

A CWE-94: Improper Control of Generation of Code ('Code Injection') vulnerability exists that allows for remote code execution when using a parameter of the DCE network settings endpoint. Affected products: StruxureWare Data Center Expert (V7.9.2 and prior)

Groovy Code Injection & SpEL Injection which lead to Remote Code Execution. This issue affected Apache ShenYu 2.4.0 and 2.4.1.

mailform version 1.04 allows remote attackers to execute arbitrary PHP code via unspecified vectors.

A Server-side Template Injection (SSTI) vulnerability exists in Nystudio107 Seomatic 3.4.12 in src/helpers/UrlHelper.php via the host header.

JIT miscompilation in the JavaScript Engine: JIT component. This vulnerability affects Firefox < 146, Firefox ESR < 115.31, Firefox ESR < 140.6, Thunderbird < 146, and Thunderbird < 140.6.

Internally used text extraction reports allow an attacker to inject code that can be executed by the application. An attacker could thereby control the behavior of the application.

In Fat-Free Framework 3.7.1, attackers can achieve arbitrary code execution if developers choose to pass user controlled input (e.g., $_REQUEST, $_GET, or $_POST) to the framework's Clear method.

A code injection vulnerability in the Ivanti EPM Cloud Services Appliance (CSA) allows an unauthenticated user to execute arbitrary code with limited permissions (nobody).

A CWE-94: Code Injection vulnerability exists in ProClima (all versions prior to version 8.0.0) which could allow an unauthenticated, remote attacker to execute arbitrary code on the targeted system in all versions of ProClima prior to version 8.0.0.

TOTOlink EX1800T V9.1.0cu.2112_B20220316 is vulnerable to unauthorized arbitrary command execution in the ‘hour’ parameter of the setRebootScheCfg interface of the cstecgi .cgi.

An issue in Blurams Lumi Security Camera (A31C) v23.0406.435.4120 allows attackers to execute arbitrary code.

Client-side printing services SAP Cloud Print Manager and SAPSprint for SAP NetWeaver Application Server for ABAP - versions 7.70, 7.70 PI, 7.70 BYD, allow an attacker to inject code that can be executed by the application. An attacker could thereby control the behavior of the application.

The PING function on the TP-Link TL-WR840N EU v5 router with firmware through TL-WR840N(EU)_V5_171211 is vulnerable to remote code execution via a crafted payload in an IP address input field.

In the SEOmatic plugin up to 3.4.11 for Craft CMS 3, it is possible for unauthenticated attackers to perform a Server-Side Template Injection, allowing for remote code execution.

CMSUno version 1.7.2 is affected by a PHP code execution vulnerability. sauvePass action in {webroot}/uno/central.php file calls to file_put_contents() function to write username in password.php file when a user successfully changed their password. The attacker can inject malicious PHP code into password.php and then use the login function to execute code.

HHEE system has a Code Injection vulnerability.Successful exploitation of this vulnerability may affect HHEE system integrity.

Cobbler before 3.3.0 allows log poisoning, and resultant Remote Code Execution, via an XMLRPC method that logs to the logfile for template injection.

opensysusers through 0.6 does not safely use eval on files in sysusers.d that may contain shell metacharacters. For example, it allows command execution via a crafted GECOS field whereas systemd-sysusers (a program with the same specification) does not do that.

playSMS before 1.4.5 allows Arbitrary Code Execution by entering PHP code at the #tabs-information-page of core_main_config, and then executing that code via the index.php?app=main&inc=core_welcome URI.

Possible Command injection Vulnerability in iManager has been discovered in OpenText™ iManager 3.2.4.0000.

Nagios XI before version 5.11.3 was discovered to contain a remote code execution (RCE) vulnerability via the component command_test.php.

A vulnerability was found in Analytics Stats Counter Statistics Plugin 1.2.2.5 and classified as critical. This issue affects some unknown processing. The manipulation leads to code injection. The attack may be initiated remotely.

An issue in NCR Terminal Handler v.1.5.1 allows a remote attacker to execute arbitrary code and obtain sensitive information via a GET request to a UserService SOAP API endpoint to validate if a user exists.

An arbitrary code execution which results in privilege escalation was discovered in Helix Core versions prior to 2023.2. Reported by Jason Geffner.

Versions of the package jsonpath before 1.2.0 are vulnerable to Arbitrary Code Injection via unsafe evaluation of user-supplied JSON Path expressions. The library relies on the static-eval module to process JSON Path input, which is not designed to handle untrusted data safely. An attacker can exploit this vulnerability by supplying a malicious JSON Path expression that, when evaluated, executes arbitrary JavaScript code, leading to Remote Code Execution in Node.js environments or Cross-site Scripting (XSS) in browser contexts. This affects all methods that evaluate JSON Paths against objects, including .query, .nodes, .paths, .value, .parent, and .apply.

vm2 is an advanced vm/sandbox for Node.js. The library contains critical security issues and should not be used for production. The maintenance of the project has been discontinued. In vm2 for versions up to 3.9.19, `Promise` handler sanitization can be bypassed with the `@@species` accessor property allowing attackers to escape the sandbox and run arbitrary code, potentially allowing remote code execution inside the context of vm2 sandbox. Version 3.10.0 contains a patch for the issue.

RCE (Remote Code Execution) vulnerability was found in some Furukawa ONU models, this vulnerability allows remote unauthenticated users to send arbitrary commands to the device via web interface.

Video Insight VMS versions prior to 7.6.1 allow remote attackers to conduct code injection attacks via unspecified vectors.

An issue discovered in phpwcms 1.9.25 allows remote attackers to run arbitrary code via DB user field during installation.

baserCMS is a website development framework. In versions 4.6.0 through 4.7.6, there is a Code Injection vulnerability in the mail form of baserCMS. As of time of publication, no known patched versions are available.

main.py in Searchor before 2.4.2 uses eval on CLI input, which may cause unexpected code execution.

A remote code execution vulnerability exists in the install functionality of OS4Ed openSIS 7.4. The username variable which is set at line 121 in install/Step5.php allows for injection of PHP code into the Data.php file that it writes. An attacker can send an HTTP request to trigger this vulnerability.

An issue in Shenzhen TCL Browser TV Web BrowseHere (aka com.tcl.browser) 6.65.022_dab24cc6_231221_gp allows a remote attacker to execute arbitrary JavaScript code via the com.tcl.browser.portal.browse.activity.BrowsePageActivity component.

A flaw was found in GNOME Maps, which is vulnerable to a code injection attack via its service.json configuration file. If the configuration file is malicious, it may execute arbitrary code.

MajorDoMo (aka Major Domestic Module) allows unauthenticated remote code execution via the admin panel's PHP console feature. An include order bug in modules/panel.class.php causes execution to continue past a redirect() call that lacks an exit statement, allowing unauthenticated requests to reach the ajax handler in inc_panel_ajax.php. The console handler within that file passes user-supplied input from GET parameters (via register_globals) directly to eval() without any authentication check. An attacker can execute arbitrary PHP code by sending a crafted GET request to /admin.php with ajax_panel, op, and command parameters.

Narou (aka Narou.rb) before 3.8.0 allows Ruby Code Injection via the title name or author name of a novel.

In Moodle, a remote code execution risk was identified in the Shibboleth authentication plugin.

The website builder module in Dolibarr 13.0.2 allows remote PHP code execution because of an incomplete protection mechanism in which system, exec, and shell_exec are blocked but backticks are not blocked.

Frauscher Sensortechnik GmbH FDS101 for FAdC/FAdCi v1.4.24 and all previous versions are vulnerable to a remote code execution (RCE) vulnerability via manipulated parameters of the web interface without authentication. This could lead to a full compromise of the FDS101 device.

vLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1 and prior to version 0.14.0, vLLM loads Hugging Face `auto_map` dynamic modules during model resolution without gating on `trust_remote_code`, allowing attacker-controlled Python code in a model repo/path to execute at server startup. An attacker who can influence the model repo/path (local directory or remote Hugging Face repo) can achieve arbitrary code execution on the vLLM host during model load. This happens before any request handling and does not require API access. Version 0.14.0 fixes the issue.

Eigent is a multi-agent Workforce. A critical security vulnerability in the CI workflow (.github/workflows/ci.yml) allows arbitrary code execution from fork pull requests with repository write permissions. The vulnerable workflow uses pull_request_target trigger combined with checkout of untrusted PR code. An attacker can exploit this to steal credentials, post comments, push code, or create releases.

An issue in FeMiner WMS v1.1 allows attackers to execute arbitrary code via the filename parameter and the exec function.

eSST Monitoring v2.147.1 was discovered to contain a remote code execution (RCE) vulnerability via the Gii code generator component.

When malicious images are pulled by isula pull, attackers can execute arbitrary code.

The TELNET service in Mobatek MobaXterm 10.4 does not require authentication, which allows remote attackers to execute arbitrary commands via TCP port 23.

A code injection in Ivanti Endpoint Manager Mobile allowing attackers to achieve unauthenticated remote code execution.

A code injection in Ivanti Endpoint Manager Mobile allowing attackers to achieve unauthenticated remote code execution.