TorchServe is a tool for serving and scaling PyTorch models in production. TorchServe default configuration lacks proper input validation, enabling third parties to invoke remote HTTP download requests and write files to the disk. This issue could be taken advantage of to compromise the integrity of the system and sensitive data. This issue is present in versions 0.1.0 to 0.8.1. A user is able to load the model of their choice from any URL that they would like to use. The user of TorchServe is responsible for configuring both the allowed_urls and specifying the model URL to be used. A pull request to warn the user when the default value for allowed_urls is used has been merged in PR #2534. TorchServe release 0.8.2 includes this change. Users are advised to upgrade. There are no known workarounds for this issue.

The Wp-Insert plugin through 2.4.2 for WordPress allows upload of arbitrary PHP code because of the exposure and configuration of FCKeditor under fckeditor/editor/filemanager/browser/default/browser.html, fckeditor/editor/filemanager/connectors/test.html, and fckeditor/editor/filemanager/connectors/uploadtest.html.

ProxyShell is an exploit chain targeting on-premise installations of Microsoft Exchange Server. It was demonstrated by Orange Tsai at Pwn2Own in April 2021 and is comprised of three CVEs that, when chained, allow a remote unauthenticated attacker to execute arbitrary code on vulnerable targets. The three CVEs are CVE-2021-34473, CVE-2021-34523, and CVE-2021-31207. Details are available in Orange Tsai's [Black Hat USA 2020 talk](https://i.blackhat.com/USA21/Wednesday-Handouts/us-21-ProxyLogon-Is-Just-The-Tip-Of-The-Iceberg-A-New-Attack-Surface-On-Microsoft-Exchange-Server.pdf) and follow-on [blog series](https://blog.orange.tw/2021/08/proxylogon-a-new-attack-surface-on-ms-exchange-part-1.html). ProxyShell is being broadly exploited in the wild as of August 12, 2021.

When Security Assertion Markup Language (SAML) authentication is enabled and the 'Validate Identity Provider Certificate' option is disabled (unchecked), improper verification of signatures in PAN-OS SAML authentication enables an unauthenticated network-based attacker to access protected resources. The attacker must have network access to the vulnerable server to exploit this vulnerability.

Adobe ColdFusion versions 2018 Update 15 (and earlier) and 2021 Update 5 (and earlier) are affected by a Deserialization of Untrusted Data vulnerability that could result in arbitrary code execution in the context of the current user. Exploitation of this issue does not require user interaction.

Fortra (formerly, HelpSystems) GoAnywhere MFT suffers from a pre-authentication command injection vulnerability in the License Response Servlet due to deserializing an arbitrary attacker-controlled object.

On WatchGuard Firebox and XTM appliances, an unauthenticated user can execute arbitrary code, aka FBX-22786. This vulnerability impacts Fireware OS before 12.7.2_U2, 12.x before 12.1.3_U8, and 12.2.x through 12.5.x before 12.5.9_U2.

Windows TCP/IP Remote Code Execution Vulnerability

smtp_mailaddr in smtp_session.c in OpenSMTPD 6.6, as used in OpenBSD 6.6 and other products, allows remote attackers to execute arbitrary commands as root via a crafted SMTP session, as demonstrated by shell metacharacters in a MAIL FROM field. This affects the "uncommented" default configuration. The issue exists because of an incorrect return value upon failure of input validation.

An issue was discovered in through SaltStack Salt before 3002.5. salt-api does not honor eauth credentials for the wheel_async client. Thus, an attacker can remotely run any wheel modules on the master.