CWE-923
Improper Restriction of Communication Channel to Intended Endpoints
Monthly
Authenticated users of Hitachi Energy TRO600 series devices can abuse the product's port-forwarding capability to reach services that are bound only to localhost, bypassing the network access restrictions that are supposed to keep those services unreachable from the network. Successful tunnelling gives the attacker access to sensitive local functionality and confidential information, and the flaw also carries an availability impact; all firmware versions from 9.0.0.0 through 9.2.1.6 are listed as affected. Exploitation requires a valid authenticated session (CVSS PR:L) and conditions outside the attacker's full control (AC:H, AT:P), no public exploit has been identified at time of analysis, and the vendor's priority guidance places this in the moderate, address-this-week band rather than an emergency.
Privilege escalation through the Azure Key Vault SDK client libraries (azure-keyvault-secrets, azure-keyvault-keys, azure-keyvault-certificates and azure-keyvault-administration) allows an authorized attacker who already holds a valid low-privilege Key Vault identity to steer the SDK's communication channel toward an unintended endpoint, exposing secrets, keys and certificates and thereby gaining the vault's data-plane privileges. Exploitation is not trivial: the assessed vector (CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:C/C:H/I:H/A:N) requires an authenticated low-privilege identity plus specialized preconditions such as attacker-influenced network positioning or precise timing, with no user interaction and no availability impact, while scope is changed because compromise of the client library reaches the Key Vault resource itself. No confirmed active exploitation (not listed in CISA KEV) and no public exploit code were identified at time of analysis, and CISA's SSVC decision framework rates Exploitation as 'none' and Automatable as 'no', although Technical Impact is rated total; Microsoft has released a vendor patch. This is a genuine but second-tier priority rather than an emergency.
Local privilege escalation in Windows Security Center lets an attacker who already holds a foothold on a Windows host with low-level privileges (CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H, 7.0 High) abuse an improperly restricted communication channel to obtain high confidentiality and integrity impact on the same machine. Only hosts that have not yet applied the current cumulative builds are affected, spanning Windows 10 1607 through 22H2, Windows 11 23H2 through 26H2, and Windows Server 2016/2019 including Server Core installations. The high attack complexity indicates a non-trivial precondition - most consistent with a race or transient-state condition that must be won to redirect or spoof the endpoint - so reliability is limited; prior local access and valid credentials are mandatory and no public exploit identified at time of analysis, with no CISA KEV listing.
StartTLS handling in 389-ds-base does not discard plaintext bytes buffered before the TLS upgrade, allowing an on-path attacker to inject a crafted LDAP message whose messageID collides with a client's pending bind and causes the client to treat a failed authentication as successful. Red Hat Directory Server 11, 12, and 13 and Red Hat Enterprise Linux 6, 7, 8, 9, and 10 deployments using 389-ds-base with StartTLS are affected. Exploitation requires an on-path network position and timing control during StartTLS negotiation, so the real-world bar is higher than the reported 9.0 base score; no public exploit identified at time of analysis.
Local privilege escalation in Meta Horizon OS prior to v66.0.0.733.524 allows a malicious on-device application to register a broadcast receiver for com.oculus.systemactivities.SCREENSHOT and receive a privileged PendingIntent containing a com.oculus.horizon CallerIdentity, which can then be used to impersonate that package to OS endpoints that rely on CallerIdentity authentication. The attack requires the malicious app to already be installed and running on the same device, via sideloading or a malicious store app; there is no remote network-only path in the authoritative assessment. No public exploit was identified at time of analysis, EPSS is low at 0.23% (13th percentile), and the available fix is Meta Horizon OS v66.0.0.733.524.
Meta Horizon OS versions prior to v74.0.0.878.1682 allow a locally installed malicious application that has been granted notification-access to impersonate the privileged com.oculus.vrshell package - and any other package signed with the same signing key - against OS endpoints that authenticate callers via CallerIdentity, producing high confidentiality and integrity impact. The root cause is an improper communication-channel restriction (CWE-923) in MediaSyncJobReceiver, which can be induced to send a privileged PendingIntent carrying a com.oculus.vrshell CallerIdentity to any application registered as a NotificationListenerService. Although the supplied NVD score is 9.1 with an AV:N vector, the realistic attack path is a local, on-device app-to-app attack that requires an attacker-controlled listener already resident on the device; EPSS is low (0.23%, 13th percentile) and no confirmed active exploitation (CISA KEV) or public exploit code was identified at time of analysis.
Remote denial of service in the utcp OCaml TCP userspace library before version 0.0.6 is possible via crafted out-of-order TCP segment sequences that expose a flaw in the segment reassembly logic. Unauthenticated network attackers can exploit this against any service built on the affected library, causing complete availability loss. No public exploit code or active exploitation has been identified at time of analysis.
libcurl's SASL state machine for LDAP authentication incorrectly treats an incomplete cryptographic handshake as a fully verified authentication exchange, allowing a Man-in-the-Middle attacker to inject a premature termination message that bypasses peer validation entirely. All curl releases from 7.82.0 through 8.21.0 are affected, spanning a multi-year release history documented in EUVD-2026-72164 and the curl project's own advisory. No public exploit code has been identified and the vulnerability is not listed in CISA KEV, though successful exploitation grants an attacker the ability to intercept LDAP credentials and inject falsified directory responses.
Privilege elevation in Microsoft Azure SQL Managed Instance lets an unauthenticated network attacker cross the intended communication-channel boundary and gain elevated privileges, carrying a scope-changed CVSS of 10.0 that reflects likely cross-tenant or cross-service impact in this multi-tenant PaaS offering. Microsoft tags it as an authentication bypass exposing high-value confidentiality and integrity, with no availability loss. There is no public exploit identified at time of analysis and EPSS probability is low (0.36%, 29th percentile), consistent with a vendor-discovered, service-side issue rather than mass exploitation.
Credential exfiltration in Amazon MQ MCP Server (awslabs.amazon-mq-mcp-server) before version 2.0.24 allows a remote unauthenticated attacker - operating through prompt injection - to redirect RabbitMQ broker credentials or OAuth access tokens to an attacker-controlled endpoint by supplying a crafted broker hostname in the MCP client context. The RabbitMQ broker connection tools fail to validate that hostnames resolved through the MCP client context correspond to legitimate Amazon MQ broker endpoints, enabling a server-side request forgery-like exfiltration primitive. No public exploit code has been identified at time of analysis, no CISA KEV listing exists, and the vendor (Amazon) has released a confirmed fix in version 2.0.24.
Authenticated users of Hitachi Energy TRO600 series devices can abuse the product's port-forwarding capability to reach services that are bound only to localhost, bypassing the network access restrictions that are supposed to keep those services unreachable from the network. Successful tunnelling gives the attacker access to sensitive local functionality and confidential information, and the flaw also carries an availability impact; all firmware versions from 9.0.0.0 through 9.2.1.6 are listed as affected. Exploitation requires a valid authenticated session (CVSS PR:L) and conditions outside the attacker's full control (AC:H, AT:P), no public exploit has been identified at time of analysis, and the vendor's priority guidance places this in the moderate, address-this-week band rather than an emergency.
Privilege escalation through the Azure Key Vault SDK client libraries (azure-keyvault-secrets, azure-keyvault-keys, azure-keyvault-certificates and azure-keyvault-administration) allows an authorized attacker who already holds a valid low-privilege Key Vault identity to steer the SDK's communication channel toward an unintended endpoint, exposing secrets, keys and certificates and thereby gaining the vault's data-plane privileges. Exploitation is not trivial: the assessed vector (CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:C/C:H/I:H/A:N) requires an authenticated low-privilege identity plus specialized preconditions such as attacker-influenced network positioning or precise timing, with no user interaction and no availability impact, while scope is changed because compromise of the client library reaches the Key Vault resource itself. No confirmed active exploitation (not listed in CISA KEV) and no public exploit code were identified at time of analysis, and CISA's SSVC decision framework rates Exploitation as 'none' and Automatable as 'no', although Technical Impact is rated total; Microsoft has released a vendor patch. This is a genuine but second-tier priority rather than an emergency.
Local privilege escalation in Windows Security Center lets an attacker who already holds a foothold on a Windows host with low-level privileges (CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H, 7.0 High) abuse an improperly restricted communication channel to obtain high confidentiality and integrity impact on the same machine. Only hosts that have not yet applied the current cumulative builds are affected, spanning Windows 10 1607 through 22H2, Windows 11 23H2 through 26H2, and Windows Server 2016/2019 including Server Core installations. The high attack complexity indicates a non-trivial precondition - most consistent with a race or transient-state condition that must be won to redirect or spoof the endpoint - so reliability is limited; prior local access and valid credentials are mandatory and no public exploit identified at time of analysis, with no CISA KEV listing.
StartTLS handling in 389-ds-base does not discard plaintext bytes buffered before the TLS upgrade, allowing an on-path attacker to inject a crafted LDAP message whose messageID collides with a client's pending bind and causes the client to treat a failed authentication as successful. Red Hat Directory Server 11, 12, and 13 and Red Hat Enterprise Linux 6, 7, 8, 9, and 10 deployments using 389-ds-base with StartTLS are affected. Exploitation requires an on-path network position and timing control during StartTLS negotiation, so the real-world bar is higher than the reported 9.0 base score; no public exploit identified at time of analysis.
Local privilege escalation in Meta Horizon OS prior to v66.0.0.733.524 allows a malicious on-device application to register a broadcast receiver for com.oculus.systemactivities.SCREENSHOT and receive a privileged PendingIntent containing a com.oculus.horizon CallerIdentity, which can then be used to impersonate that package to OS endpoints that rely on CallerIdentity authentication. The attack requires the malicious app to already be installed and running on the same device, via sideloading or a malicious store app; there is no remote network-only path in the authoritative assessment. No public exploit was identified at time of analysis, EPSS is low at 0.23% (13th percentile), and the available fix is Meta Horizon OS v66.0.0.733.524.
Meta Horizon OS versions prior to v74.0.0.878.1682 allow a locally installed malicious application that has been granted notification-access to impersonate the privileged com.oculus.vrshell package - and any other package signed with the same signing key - against OS endpoints that authenticate callers via CallerIdentity, producing high confidentiality and integrity impact. The root cause is an improper communication-channel restriction (CWE-923) in MediaSyncJobReceiver, which can be induced to send a privileged PendingIntent carrying a com.oculus.vrshell CallerIdentity to any application registered as a NotificationListenerService. Although the supplied NVD score is 9.1 with an AV:N vector, the realistic attack path is a local, on-device app-to-app attack that requires an attacker-controlled listener already resident on the device; EPSS is low (0.23%, 13th percentile) and no confirmed active exploitation (CISA KEV) or public exploit code was identified at time of analysis.
Remote denial of service in the utcp OCaml TCP userspace library before version 0.0.6 is possible via crafted out-of-order TCP segment sequences that expose a flaw in the segment reassembly logic. Unauthenticated network attackers can exploit this against any service built on the affected library, causing complete availability loss. No public exploit code or active exploitation has been identified at time of analysis.
libcurl's SASL state machine for LDAP authentication incorrectly treats an incomplete cryptographic handshake as a fully verified authentication exchange, allowing a Man-in-the-Middle attacker to inject a premature termination message that bypasses peer validation entirely. All curl releases from 7.82.0 through 8.21.0 are affected, spanning a multi-year release history documented in EUVD-2026-72164 and the curl project's own advisory. No public exploit code has been identified and the vulnerability is not listed in CISA KEV, though successful exploitation grants an attacker the ability to intercept LDAP credentials and inject falsified directory responses.
Privilege elevation in Microsoft Azure SQL Managed Instance lets an unauthenticated network attacker cross the intended communication-channel boundary and gain elevated privileges, carrying a scope-changed CVSS of 10.0 that reflects likely cross-tenant or cross-service impact in this multi-tenant PaaS offering. Microsoft tags it as an authentication bypass exposing high-value confidentiality and integrity, with no availability loss. There is no public exploit identified at time of analysis and EPSS probability is low (0.36%, 29th percentile), consistent with a vendor-discovered, service-side issue rather than mass exploitation.
Credential exfiltration in Amazon MQ MCP Server (awslabs.amazon-mq-mcp-server) before version 2.0.24 allows a remote unauthenticated attacker - operating through prompt injection - to redirect RabbitMQ broker credentials or OAuth access tokens to an attacker-controlled endpoint by supplying a crafted broker hostname in the MCP client context. The RabbitMQ broker connection tools fail to validate that hostnames resolved through the MCP client context correspond to legitimate Amazon MQ broker endpoints, enabling a server-side request forgery-like exfiltration primitive. No public exploit code has been identified at time of analysis, no CISA KEV listing exists, and the vendor (Amazon) has released a confirmed fix in version 2.0.24.