(A)I Sees What You Don't: Exploiting New Attack Surfaces in Third-Party Mobile Agents

Third-party mobile agents powered by Vision-Language Models VLMs have emerged as a promising paradigm for automating smartphone interactions. These agents act as high-privilege decision-makers, perceiving device states through screenshots and executing actions via VLM reasoning, transforming how ...

CVE-2026-56808: Improper Neutralization of Special Elements used in an OS Command

DGM3103SCT provided by AVTECH Security Corporation contains an OS command injection vulnerability, which may lead to arbitrary command execution with the root privilege by a user who can log in to the web management console of the affected product...

PT-2026-53949

Name of the Vulnerable Software and Affected Versions IBM Langflow OSS versions 1.0.0 through 1.10.0 Description Authenticated attackers can execute arbitrary OS commands and read sensitive files, including credentials. This can lead to complete system compromise and lateral movement within the...

RHEL 8 : vim (RHSA-2026:33453)

The remote Redhat Enterprise Linux 8 host has packages installed that are affected by multiple vulnerabilities as referenced in the RHSA-2026:33453 advisory. Vim Vi IMproved is an updated and improved version of the vi editor. Security Fixes: vim: arbitrary command execution via modeline sandbox...

PraisonAI vulnerable to sandbox escape via `print.__self__` builtins module leak in `execute_code` (subprocess mode)

Summaryexecutecode in praisonaiagents/tools/pythontools.py v1.6.37, subprocess sandbox mode can be fully bypassed using print.self to retrieve the real Python builtins module, from which import can be extracted via vars and runtime string construction. This achieves arbitrary OS command execution...

PraisonAI vulnerable to sandbox escape via `print.__self__` builtins module leak in `execute_code` (subprocess mode)

Summaryexecutecode in praisonaiagents/tools/pythontools.py v1.6.37, subprocess sandbox mode can be fully bypassed using print.self to retrieve the real Python builtins module, from which import can be extracted via vars and runtime string construction. This achieves arbitrary OS command execution...

PYSEC-2026-463 PraisonAI vulnerable to sandbox escape via `print.__self__` builtins module leak in `execute_code` (subprocess mode)

Summary executecode in praisonaiagents/tools/pythontools.py v1.6.37, subprocess sandbox mode can be fully bypassed using print.self to retrieve the real Python builtins module, from which import can be extracted via vars and runtime string construction. This achieves arbitrary OS command executio...

pymetasploit3 vulnerable to command injection in console.run_module_with_output()

Command injection vulnerability in console.runmodulewithoutput in pymetasploit3 through version 1.0.6 allows attackers to inject newline characters into module options such as RHOSTS. This breaks the intended command structure and causes the Metasploit console to execute additional unintended...

vim: arbitrary command execution via modeline sandbox bypass

A flaw was found in Vim. A modeline is used to set specific editor options directly from a text file. However, the complete, guitabtooltip, printheader options and the mapset function lack proper security checks, allowing an attacker to bypass restrictions and cause arbitrary OS command execution...

Important: Red Hat Security Advisory: vim security update

An update for vim is now available for Red Hat Enterprise Linux 10.0 Extended Update Support. Red Hat Product Security has rated this update as having a security impact of Important. A Common Vulnerability Scoring System CVSS base score, which gives a detailed severity rating, is available for ea...

CVE-2026-57919: Incorrect Default Permissions

PBackupVSS.exe in Matrix42 Empirum before 25.5 and 26.x before 26.2 creates a named pipe \.\pipe\PBackupVSS with a DACL that grants GENERICREAD and GENERICWRITE permissions to all authenticated users. A low-privileged local attacker can connect to this pipe and send crafted IPC messages to trigge...

PT-2026-53398

Summary 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. Vulnerable...

PT-2026-53613

Summary The log file name parameter in the stata do API and CLI is directly interpolated into a Stata command string without sanitization. The security guard GuardValidator only scans the do-file content but does not validate this parameter. An attacker can inject arbitrary Stata commands includi...

PT-2026-53632

The package libvcs before 0.11.1 are vulnerable to Command Injection via argument injection. When calling the update repo function when using hg, the url parameter is passed to the hg clone command. By injecting some hg options it was possible to get arbitrary command execution...

PT-2026-53361

An issue in the Pickle Python library of NASA AIT-Core v2.5.2 allows attackers to execute arbitrary commands...

PT-2026-57013

Name of the Vulnerable Software and Affected Versions Vim versions prior to 9.2.0736 Description The PHP omni-completion script in runtime/autoload/phpcomplete.vim fails to escape class or trait names taken from the edited buffer before interpolating them into a search pattern executed via win...

CVE-2026-48800 Notepad++: Arbitrary Code Execution via shortcuts.xml UserCommand Injection

Notepad++ is a free and open-source source code editor. Prior to 8.9.6.1, the tag text content inside in shortcuts.xml is read by NppXml::valueaNode Parameters.cpp:3658 in the feedUserCmds function and stored in UserCommand.cmd without any validation. When the user clicks the corresponding entry ...

Arbitrary Argument Injection

Overview @cyclonedx/cdxgen is a Creates CycloneDX Software Bill of Materials SBOM from source or container image Affected versions of this package are vulnerable to Arbitrary Argument Injection via the Maven project scanning process. An attacker can execute arbitrary shell commands by submitting ...

Incus has a restricted project bypass leading to arbitrary command execution

Summary Instance snapshots ignore the restricted.containers.lowlevel=block setting; allowing for arbitrary command execution on the Incus server by abusing lowlevel hooks such as raw.lxc and raw.qemu. Details Instance snapshots ignore the restricted.containers.lowlevel=block setting; allowing for...

GHSA-48Q5-W887-33WV Incus has a restricted project bypass leading to arbitrary command execution

Summary Instance snapshots ignore the restricted.containers.lowlevel=block setting; allowing for arbitrary command execution on the Incus server by abusing lowlevel hooks such as raw.lxc and raw.qemu. Details Instance snapshots ignore the restricted.containers.lowlevel=block setting; allowing for...