CISA adds recently disclosed Apple flaws to its Known Exploited Vulnerabilities catalog

Roundcube Webmail versions before 1.4.4 contain an OS command injection flaw (CWE-78) in rcube_image.php: shell metacharacters in the im_convert_path or im_identify_path configuration settings are not escaped before the configured ImageMagick binaries are executed. When image processing is triggered, an attacker who can control those configuration values can append arbitrary shell commands that run with the privileges of the web server user, yielding full remote code execution on the mail server. Successful exploitation can expose stored email, mail credentials, and the underlying host; the flaw is scored 9.8 (critical), with no privileges or user interaction required per the CVSS vector. Any self-hosted Roundcube deployment older than 1.4.4 is affected, including Roundcube packages shipped by openSUSE Leap and openSUSE Backports for SUSE Linux Enterprise. Exploitation is confirmed in the wild: CISA added it to the KEV catalog on 2023-06-22, EPSS puts the 30-day exploitation probability at ~84%, and headlines note APT28-linked activity targeting government Roundcube servers, including Ukrainian entities.

Do: Upgrade Roundcube to 1.4.4 or later, or apply the vendor-patched openSUSE Leap / SLE backport packages, as required by the CISA KEV listing (added 2023-06-22). After patching, verify that im_convert_path and im_identify_path settings contain no unescaped metacharacters and review web server logs for injected command activity; given APT28's targeting of government Roundcube servers, prioritize public-sector mail infrastructure for patching and threat hunting.

Roundcube Webmail contains a cross-site scripting (XSS) flaw (CWE-79) in the link-reference handling of rcube_string_replacer.php, where the linkref_addindex function mishandles JavaScript embedded in a link element of a plain-text email. An attacker triggers the flaw simply by sending a crafted plain-text message to a victim; when the message is processed/displayed in the Roundcube interface, the embedded script executes in the context of the victim's webmail session. Successful exploitation can lead to session hijacking, theft of webmail cookies or credentials, and arbitrary actions in the victim's mailbox. Any deployment of Roundcube Webmail is affected, which includes self-hosted instances and webmail offered by hosting providers, ISPs, and universities. Although no public proof-of-concept is known, CISA added this vulnerability to the Known Exploited Vulnerabilities catalog on 2023-06-22, confirming exploitation in the wild; ransomware association is unknown, and no CVSS score is yet available, though EPSS puts 30-day exploitation probability at 32.7% (98th percentile).

Do: Apply the vendor's updated Roundcube release per CISA's required action (updates per vendor instructions); since no specific fixed versions appear in this data, install the latest patched release of your deployed 1.x branch and verify with the vendor advisory. Check webmail servers for processing of plain-text messages with link-reference elements and review logs for anomalous webmail sessions; treat KEV-listed status as evidence of active exploitation and prioritize internet-exposed Roundcube instances.

Roundcube Webmail contains a SQL injection flaw (CWE-89) in which attacker-controlled 'search' or 'search_params' input is incorporated into database queries without sufficient sanitization. An attacker with access to the webmail search functionality (typically an authenticated mailbox user) can submit crafted parameters to execute arbitrary SQL against the Roundcube backend database, potentially reading or modifying stored mail account data. Any organization running Roundcube is affected, including self-hosted mail servers and customers of hosting providers that ship Roundcube as their bundled webmail client. The flaw was added to CISA's Known Exploited Vulnerabilities catalog on 2023-06-22, indicating exploitation in the wild, and EPSS assigns a 41.9% probability of exploitation within 30 days (99th percentile). No public proof-of-concept is known, and CISA lists ransomware use as unknown.

Do: Apply vendor updates per Roundcube's instructions by upgrading to the latest patched, supported release, prioritizing internet-facing webmail servers; federal agencies must remediate per the CISA KEV requirement. If Roundcube is managed by a hosting provider (e.g., via cPanel), coordinate patching with them. In the interim, restrict webmail exposure and review database and web logs for anomalous search-related queries that may indicate exploitation.

VMware Tools, the agent installed inside guest virtual machines, fails to properly authenticate host-to-guest operations when they are issued from an ESXi host (CVE-2023-20867, CWE-287 improper authentication). An attacker who has already gained full (root) control of an ESXi host can invoke these operations, such as running commands or moving files inside guest VMs, and the guests' VMware Tools will accept them without valid authentication. This gives an attacker a foothold in guest VMs without guest credentials, affecting guest confidentiality and integrity. Any organization running VMware ESXi/vSphere with VMware Tools in its guests is affected, and the component is also shipped as open-vm-tools in Debian and Fedora. The flaw is confirmed exploited in the wild: CISA added it to the Known Exploited Vulnerabilities catalog on 2023-06-23 (EPSS 13.5%, 96th percentile), and China-linked APT UNC3886, whose 'Fire Ant' tooling targets ESXi and vCenter, has been reported using it alongside other VMware flaws.

Do: Upgrade VMware Tools / open-vm-tools to the latest fixed release distributed by VMware, Debian, or Fedora per the vendor advisory, and inventory guests running outdated Tools. Because exploitation requires a fully compromised ESXi host, hunt for signs of host compromise (unexpected processes, modified VIBs, suspicious vCenter activity) and review guest VMs for unexplained command execution or persistence. Consistent with the KEV required action, prioritize patching, starting with internet-facing ESXi hosts and virtualization management infrastructure.

VMware Aria Operations for Networks (formerly vRealize Network Insight) contains a command injection flaw (CWE-77) that allows an attacker with network access to the appliance to run arbitrary operating-system commands. Because the attack requires no authentication, privileges, or user interaction (CVSS:3.1/AV:N/AC:L/PR:N/UI:N), any party able to reach the product's network interface can trigger it, gaining remote code execution with high impact on confidentiality, integrity, and availability. Any organization running the product is affected, with internet-facing deployments at the greatest risk. Exploitation is confirmed in the wild — CISA added the flaw to the KEV catalog on 2023-06-22, a public proof-of-concept exploit is available, and EPSS puts the 30-day exploitation probability at 98.3% (top percentile). Ransomware use is currently unknown.

Do: Apply the vendor's patched update to all Aria Operations for Networks deployments as soon as possible — this is also CISA's required KEV action (apply updates per vendor instructions); verify the installed build against VMware's advisory for affected ranges. Until patching is complete, restrict network access to the appliance (firewall rules, VPN, or management-segment isolation), prioritizing any instance reachable from the internet since no authentication is required to exploit. Hunt for indicators of command injection exploitation, as in-the-wild exploitation has been confirmed.

Zyxel NAS326, NAS540, and NAS542 network-attached storage devices contain a pre-authentication command injection flaw (CWE-78) that lets an unauthenticated attacker execute operating system commands by sending a crafted HTTP request to the device. Because the flaw is network-facing and requires no credentials or user interaction, a remote attacker gains the ability to run arbitrary OS commands on the device, effectively full compromise. Affected firmware is NAS326 versions prior to V5.21(AAZF.14)C0, NAS540 versions prior to V5.21(AATB.11)C0, and NAS542 versions prior to V5.21(ABAG.11)C0. The flaw scores 9.8 (critical) on CVSS 3.1, carries a very high 83.8% probability of exploitation within 30 days per EPSS, and CISA added it to the Known Exploited Vulnerabilities catalog on 2023-06-23. News reporting indicates a Mirai-like botnet is already exploiting the flaw in the wild, so defenders should treat it as an actively exploited, internet-exposable issue and patch immediately.

Do: Upgrade NAS326 to V5.21(AAZF.14)C0, NAS540 to V5.21(AATB.11)C0, and NAS542 to V5.21(ABAG.11)C0 per Zyxel's security advisories. Until patched, keep the NAS web administration interface off the public internet or restrict access with firewall rules. Because a Mirai-like botnet is actively exploiting this flaw, inspect patched and unpatched devices for signs of compromise, such as unfamiliar processes or unexpected outbound traffic; organizations covered by CISA's KEV requirements must apply the vendor updates by the required deadline.

Apple's WebKit engine, used by Safari and shipped with iOS, iPadOS, and macOS, contains a type confusion vulnerability (CWE-843) that leads to code execution when processing maliciously crafted web content. An attacker can trigger the flaw by inducing a victim to load attacker-controlled web content, for example by visiting a crafted webpage or opening malicious HTML in any application that renders it with WebKit. Successful exploitation grants the attacker arbitrary code execution within the web-content processing context of the affected browser or application. The exposure is broad: every iPhone, iPad, and Mac running vulnerable WebKit builds is affected, and the flaw could also impact HTML parsers in non-Apple products that rely on WebKit. The vulnerability was added to CISA's Known Exploited Vulnerabilities catalog on 2023-06-23, confirming known in-the-wild exploitation (ransomware use unknown); no public proof-of-concept is known, and CISA's required action is to apply updates per vendor instructions.

Do: Apply Apple's security updates for iOS, iPadOS, macOS, and Safari immediately, following vendor instructions as required by the CISA KEV catalog, since the flaw is confirmed exploited in the wild. Administrators should also inventory any non-Apple applications, HTML parsers, or embedded browsers that use WebKit and apply the corresponding vendor patches when available. Until patching is complete, treat unsolicited web links and HTML content as a primary attack vector and prioritize updates on internet-facing and user-facing Apple systems.