PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the workflow engine is vulnerable to arbitrary command and code execution through untrusted YAML files. When praisonai workflow run <file.yaml> loads a YAML file with type: job, the JobWorkflowExecutor in job_workflow.py processes steps that support run: (shell commands via subprocess.run()), script: (inline Python via exec()), and python: (arbitrary Python script execution)—all without any validation, sandboxing, or user confirmation. The affected code paths include action_run() in workflow.py and _exec_shell(), _exec_inline_python(), and _exec_python_script() in job_workflow.py. An attacker who can supply or influence a workflow YAML file (particularly in CI pipelines, shared repositories, or multi-tenant deployment environments) can achieve full arbitrary command execution on the host system, compromising the machine and any accessible data or credentials. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents.

PraisonAI is a multi-agent teams system. Prior to version 1.5.90, run_python() in praisonai constructs a shell command string by interpolating user-controlled code into python3 -c "<code>" and passing it to subprocess.run(..., shell=True). The escaping logic only handles \ and ", leaving $() and backtick substitutions unescaped, allowing arbitrary OS command execution before Python is invoked. This issue has been patched in version 1.5.90.

PraisonAI is a multi-agent teams system. Prior to 4.5.115, the AgentService.loadAgentFromFile method uses the js-yaml library to parse YAML files without disabling dangerous tags (such as !!js/function and !!js/undefined). This allows an attacker to craft a malicious YAML file that, when parsed, executes arbitrary JavaScript code. An attacker can exploit this vulnerability by uploading a malicious agent definition file via the API endpoint, leading to remote code execution (RCE) on the server. This vulnerability is fixed in 4.5.115.

PraisonAI is a multi-agent teams system. Prior to version 4.5.90, the get_all_user_threads function constructs raw SQL queries using f-strings with unescaped thread IDs fetched from the database. An attacker stores a malicious thread ID via update_thread. When the application loads the thread list, the injected payload executes and grants full database access. This issue has been patched in version 4.5.90.

PraisonAI is a multi-agent teams system. Prior to 4.5.121, the execute_command function and workflow shell execution are exposed to user-controlled input via agent workflows, YAML definitions, and LLM-generated tool calls, allowing attackers to inject arbitrary shell commands through shell metacharacters. This vulnerability is fixed in 4.5.121.

PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, he memory hooks executor in praisonaiagents passes a user-controlled command string directly to subprocess.run() with shell=True at src/praisonai-agents/praisonaiagents/memory/hooks.py. No sanitization is performed and shell metacharacters are interpreted by /bin/sh before the intended command executes. Two independent attack surfaces exist. The first is via pre_run_command and post_run_command hook event types registered through the hooks configuration. The second and more severe surface is the .praisonai/hooks.json lifecycle configuration, where hooks registered for events such as BEFORE_TOOL and AFTER_TOOL fire automatically during agent operation. An agent that gains file-write access through prompt injection can overwrite .praisonai/hooks.json and have its payload execute silently at every subsequent lifecycle event without further user interaction. This vulnerability is fixed in 1.5.128.

PraisonAI is a multi-agent teams system. Prior to version 4.5.97, SubprocessSandbox in all modes (BASIC, STRICT, NETWORK_ISOLATED) calls subprocess.run() with shell=True and relies solely on string-pattern matching to block dangerous commands. The blocklist does not include sh or bash as standalone executables, allowing trivial sandbox escape in STRICT mode via sh -c '<command>'. This issue has been patched in version 4.5.97.

X-Plane before 11.41 has multiple improper path validations that could allow reading and writing files from/to arbitrary paths (or a leak of OS credentials to a remote system) via crafted network packets. This could be used to execute arbitrary commands on the system.

# Vulnerability in `rand-quote` and `hitokoto` plugins **Description**: the `rand-quote` and `hitokoto` fetch quotes from quotationspage.com and hitokoto.cn respectively, do some process on them and then use `print -P` to print them. If these quotes contained the proper symbols, they could trigger command injection. Given that they're an external API, it's not possible to know if the quotes are safe to use. **Fixed in**: [72928432](https://github.com/ohmyzsh/ohmyzsh/commit/72928432). **Impacted areas**: - `rand-quote` plugin (`quote` function). - `hitokoto` plugin (`hitokoto` function).

emfd in Ruckus Wireless Unleashed through 200.7.10.102.64 allows remote attackers to execute OS commands via a POST request with the attribute xcmd=import-category to admin/_cmdstat.jsp via the uploadFile attribute.

A security vulnerability has been detected in Totolink A7100RU 7.4cu.2313_b20191024. Affected is the function setWiFiAclRules of the file /cgi-bin/cstecgi.cgi of the component CGI Handler. The manipulation of the argument mode leads to os command injection. The attack can be initiated remotely. The exploit has been disclosed publicly and may be used.

emfd in Ruckus Wireless Unleashed through 200.7.10.102.64 allows remote attackers to execute OS commands via a POST request with the attribute xcmd=spectra-analysis to admin/_cmdstat.jsp via the mac attribute.

SOUND4 IMPACT/FIRST/PULSE/Eco versions 2.x and below contain an unauthenticated command injection vulnerability in the username parameter. Attackers can exploit index.php and login.php scripts by injecting arbitrary shell commands through the HTTP POST 'username' parameter to execute system commands.

A weakness has been identified in Totolink A7100RU 7.4cu.2313_b20191024. This issue affects the function setUrlFilterRules of the file /cgi-bin/cstecgi.cgi of the component CGI Handler. Executing a manipulation of the argument enable can lead to os command injection. The attack can be launched remotely. The exploit has been made available to the public and could be used for attacks.

emfd in Ruckus Wireless Unleashed through 200.7.10.102.64 allows remote attackers to execute OS commands via a POST request with the attribute xcmd=packet-capture to admin/_cmdstat.jsp via the mac attribute.

A vulnerability was found in Totolink A7100RU 7.4cu.2313_b20191024. Affected by this vulnerability is the function setTracerouteCfg of the file /cgi-bin/cstecgi.cgi of the component CGI Handler. The manipulation of the argument command results in os command injection. The attack may be launched remotely. The exploit has been made public and could be used.

Tdarr 2.00.15 contains an unauthenticated remote code execution vulnerability in its Help terminal that allows attackers to inject and chain arbitrary commands. Attackers can exploit the lack of input filtering by chaining commands like `--help; curl .py | python` to execute remote code without authentication.

A flaw has been found in Totolink A7100RU 7.4cu.2313_b20191024. This affects the function setAppCfg of the file /cgi-bin/cstecgi.cgi of the component CGI Handler. Executing a manipulation of the argument enable can lead to os command injection. The attack may be launched remotely. The exploit has been published and may be used.

Tellabs Optical Line Terminal (OLT) 1150 devices allow Remote Command Execution via the -l option to TELNET or SSH. Tellabs has addressed this issue in the SR30.1 and SR31.1 release on February 18, 2020.

A security vulnerability has been detected in Totolink A7100RU 7.4cu.2313_b20191024. Affected by this vulnerability is the function setIpv6LanCfg of the file /cgi-bin/cstecgi.cgi of the component CGI Handler. Such manipulation of the argument addrPrefixLen leads to os command injection. The attack may be performed from remote. The exploit has been disclosed publicly and may be used.

A security vulnerability has been detected in Totolink A7100RU 7.4cu.2313_b20191024. The affected element is the function setAdvancedInfoShow of the file /cgi-bin/cstecgi.cgi of the component CGI Handler. The manipulation of the argument tty_server leads to os command injection. It is possible to initiate the attack remotely. The exploit has been disclosed publicly and may be used.

emfd in Ruckus Wireless Unleashed through 200.7.10.102.64 allows remote attackers to execute OS commands via a POST request with the attribute xcmd=get-platform-depends to admin/_cmdstat.jsp via the uploadFile attribute.

D-Link DIR-846 devices with firmware 100A35 allow remote attackers to execute arbitrary OS commands as root by leveraging admin access and sending a /HNAP1/ request for SetWizardConfig with shell metacharacters to /squashfs-root/www/HNAP1/control/SetWizardConfig.php.

D-Link DIR-1260 Wi-Fi router firmware versions up to and including v1.20B05 contain a command injection vulnerability within the web management interface that allows for unauthenticated attackers to execute arbitrary commands on the device with root privileges. The flaw specifically exists within the SetDest/Dest/Target arguments to the GetDeviceSettings form. The management interface is accessible over HTTP and HTTPS on the local and Wi-Fi networks and optionally from the Internet.

Unauthenticated remote code execution occurs in D-Link products such as DIR-655C, DIR-866L, DIR-652, and DHP-1565. The issue occurs when the attacker sends an arbitrary input to a "PingTest" device common gateway interface that could lead to common injection. An attacker who successfully triggers the command injection could achieve full system compromise. Later, it was independently found that these are also affected: DIR-855L, DAP-1533, DIR-862L, DIR-615, DIR-835, and DIR-825.

On D-Link DIR-859 A3-1.06 and DIR-850 A1.13 devices, /etc/services/DEVICE.TIME.php allows command injection via the $SERVER variable.

A OS Command Injection vulnerability in the bootstrap stage of Bitdefender BOX 2 allows the manipulation of the `get_image_url()` function in special circumstances to inject a system command.

A vulnerability was determined in Totolink A7100RU 7.4cu.2313_b20191024. Affected by this issue is the function setLedCfg of the file /cgi-bin/cstecgi.cgi of the component CGI Handler. This manipulation of the argument enable causes os command injection. Remote exploitation of the attack is possible. The exploit has been publicly disclosed and may be utilized.

The UPnP endpoint URL /gena.cgi in the D-Link DIR-859 Wi-Fi router 1.05 and 1.06B01 Beta01 allows an Unauthenticated remote attacker to execute system commands as root, by sending a specially crafted HTTP SUBSCRIBE request to the UPnP service when connecting to the local network.

The processCommandUploadLog() function of libcommon.so in Petwant PF-103 firmware 4.22.2.42 and Petalk AI 3.2.2.30 allows remote attackers to execute arbitrary system commands as the root user.

A command injection vulnerability has been discovered in the bootstrap stage of Bitdefender BOX 2, versions 2.1.47.42 and 2.1.53.45. The API method `/api/download_image` unsafely handles the production firmware URL supplied by remote servers, leading to arbitrary execution of system commands. In order to exploit the condition, an unauthenticated attacker should impersonate a infrastructure server to trigger this vulnerability.

Crestron DMC-STRO 1.0 devices allow remote command execution as root via shell metacharacters to the ping function.

A shell injection vulnerability on the Sophos Cyberoam firewall appliance with CyberoamOS before 10.6.6 MR-6 allows remote attackers to execute arbitrary commands via the Web Admin and SSL VPN consoles.

D-Link DIR-846 devices with firmware 100A35 allow remote attackers to execute arbitrary OS commands as root by leveraging admin access and sending a /HNAP1/ request for SetMasterWLanSettings with shell metacharacters to /squashfs-root/www/HNAP1/control/SetMasterWLanSettings.php.

MiniDVBLinux 5.4 contains a remote command execution vulnerability that allows unauthenticated attackers to execute arbitrary commands as root through the 'command' GET parameter. Attackers can exploit the /tpl/commands.sh endpoint by sending malicious command values to gain root-level system access.

An issue was discovered on Xiaomi Mi WiFi R3G devices before 2.28.23-stable. The backup file is in tar.gz format. After uploading, the application uses the tar zxf command to decompress, so one can control the contents of the files in the decompressed directory. In addition, the application's sh script for testing upload and download speeds reads a URL list from /tmp/speedtest_urls.xml, and there is a command injection vulnerability, as demonstrated by api/xqnetdetect/netspeed.

An issue was discovered in SageMath Sage Cell Server through 2019-10-05. Python Code Injection can occur in the context of an internet facing web application. Malicious actors can execute arbitrary commands on the underlying operating system, as demonstrated by an __import__('os').popen('whoami').read() line. NOTE: the vendor's position is that the product is "vulnerable by design" and the current behavior will be retained

Intellian Remote Access 3.18 allows remote attackers to execute arbitrary OS commands via shell metacharacters in the Ping Test field.

pacman before 5.2 is vulnerable to arbitrary command injection in conf.c in the download_with_xfercommand() function. This can be exploited when unsigned databases are used. To exploit the vulnerability, the user must enable a non-default XferCommand and retrieve an attacker-controlled crafted database and package.

A remote code execution vulnerability in the PAN-OS SSH device management interface that can lead to unauthenticated remote users with network access to the SSH management interface gaining root access to PAN-OS. This issue affects PAN-OS 7.1 versions prior to 7.1.24-h1, 7.1.25; 8.0 versions prior to 8.0.19-h1, 8.0.20; 8.1 versions prior to 8.1.9-h4, 8.1.10; 9.0 versions prior to 9.0.3-h3, 9.0.4.

processCommandSetUid() in libcommon.so in Petwant PF-103 firmware 4.22.2.42 and Petalk AI 3.2.2.30 allows remote attackers to execute arbitrary system commands as the root user.

The login_mgr.cgi script in D-Link DNS-320 through 2.05.B10 is vulnerable to remote command injection.

SITOS six Build v6.2.1 allows an attacker to inject arbitrary PHP commands. As a result, an attacker can compromise the running server and execute system commands in the context of the web user.

A vulnerability was detected in Totolink A7100RU 7.4cu.2313_b20191024. The affected element is the function setVpnAccountCfg of the file /cgi-bin/cstecgi.cgi of the component CGI Handler. The manipulation of the argument User results in os command injection. The attack may be launched remotely. The exploit is now public and may be used.

processCommandUpgrade() in libcommon.so in Petwant PF-103 firmware 4.22.2.42 and Petalk AI 3.2.2.30 allows remote attackers to execute arbitrary system commands as the root user.

The processCommandSetMac() function of libcommon.so in Petwant PF-103 firmware 4.22.2.42 and Petalk AI 3.2.2.30 allows remote attackers to execute arbitrary system commands as the root user.

ALLNET ALL-RUT22GW v3.3.8 was discovered to contain an OS command injection vulnerability via the command parameter in the popen.cgi endpoint.

An OS command injection vulnerability in the discover_and_manage CGI script in NETSAS Enigma NMS 65.0.0 and prior allows an attacker to execute arbitrary code because of improper neutralization of shell metacharacters in the ip_address variable within an snmp_browser action.

An issue was found on the Ruijie EG-2000 series gateway. There is a newcli.php API interface without access control, which can allow an attacker (who only has web interface access) to use TELNET commands and/or show admin passwords via the mode_url=exec&command= substring. This affects EG-2000SE EG_RGOS 11.9 B11P1.

An issue in dlink DIR 823x 240802 allows a remote attacker to execute arbitrary code via the target_addr key value and the function 0x41737c