Information Disclosure

Denial of service in Linux kernel counter driver (rz-mtu3-cnt) allows local attackers with low privileges to crash the system by exploiting a race condition where counter and PWM sub-drivers overwrite a shared device pointer, causing incorrect runtime power management operations. The vulnerability affects kernel versions prior to specific patch levels across the 6.x and 7.x branches, with EPSS exploitation probability of 0.02% indicating low real-world exploitation likelihood despite the availability of a vendor patch.

Denial of service in Linux kernel VXLAN module allows local authenticated attackers to crash the system via malformed IPv6 neighbor discovery options in vxlan_na_create(). A crafted ND option with incorrect length values can cause out-of-bounds access or undersized payload reads, triggering a kernel panic. EPSS exploitation probability is low at 0.02%, but the vulnerability is confirmed patched across multiple stable kernel branches (5.10.253, 5.15.203, 6.1.168, 6.6.134, 6.12.81, 6.18.22, 6.19.12).

Memory leaks in the ftgmac100 Ethernet driver's ring allocation function allow local authenticated users to cause a denial of service through resource exhaustion on driver initialization failure. The vulnerability is triggered when ftgmac100_alloc_rings() fails during intermediate allocation stages, returning directly without freeing previously allocated resources (rx_skbs, tx_skbs, rxdes, txdes, rx_scratch). This affects Linux kernel versions prior to fixes released in stable branches 5.10.253, 5.15.203, 6.1.168, 6.6.134, 6.12.81, 6.18.22, 6.19.12, and 7.0.

Denial of service in Linux kernel scheduler extension (sched_ext) allows local privileged attackers to crash systems by triggering incorrect task migration validation. The vulnerability exists in the is_bpf_migration_disabled() function, which fails to correctly identify migration-disabled tasks on non-PREEMPT_RCU configurations, potentially dispatching such tasks to remote CPUs and triggering kernel errors in task_can_run_on_remote_rq(). EPSS exploitation probability is very low at 0.02%, but CVSS 5.5 indicates local attackers with standard user privileges can cause denial of service.

Resource leaks in the Linux kernel GPIO subsystem allow local attackers with low privileges to cause denial of service through memory exhaustion during gpiochip initialization error handling. The vulnerability arises from improper reference counting in gpiochip_add_data_with_key() after device initialization, where error paths fail to release device references, leading to memory not being freed on function failure. CVSS 5.5 with EPSS 0.02% indicates low real-world exploitation probability despite local attack vector.

Denial of service in the Linux kernel USB gadget ECM (Ethernet Control Model) driver allows local authenticated attackers to crash the system by exploiting improper net_device lifecycle management during bind and unbind cycles. When the gadget device unbinds, the network device survives with dangling sysfs symlinks, causing kernel issues when accessed or when the device tree is traversed. The vulnerability affects Linux kernel versions prior to patches released in 6.12.81, 6.18.22, 6.19.12, and 7.0.

Denial of service in Linux kernel USB gadget EEM function allows local privileged attackers to crash the system by triggering a dangling sysfs symlink condition during gadget device unbind cycles. The vulnerability arises from improper net_device lifecycle management when the parent gadget device is destroyed while the network device persists, resulting in kernel panic or system instability. CVSS 5.5 reflects local privilege requirement (PR:L) and high availability impact, with EPSS at 0.02% percentile indicating minimal real-world exploitation probability despite patch availability.

USB gadget subsystem net_device lifecycle mismanagement in Linux kernel allows local privileged users to cause denial of service through sysfs corruption. The f_subset gadget function creates dangling sysfs symlinks when unbinding due to improper device reparenting, resulting in inaccessible network device references and potential system instability. A local user with sufficient privileges can trigger unbind/rebind cycles to exhaust resources or corrupt the sysfs filesystem state.

Dangling sysfs symlinks in the Linux kernel's USB gadget f_rndis driver cause availability loss on systems where the RNDIS USB gadget function is used. The net_device object outlives its sysfs parent when the USB gadget unbinds, leaving broken references under /sys/class/net/ that persist until the kernel object is reclaimed. Exploitation is local and requires low-privilege access; no public exploit or active exploitation (CISA KEV) has been identified, and EPSS probability stands at 0.02% (5th percentile), reflecting minimal real-world exploitation interest at time of analysis.

List corruption in the Linux kernel's USB HID gadget driver (f_hid) can crash systems configured with USB gadget functionality. When a user-space process opens /dev/hidg0, registers the file descriptor with epoll, and the UDC (USB Device Controller) is subsequently unbound and rebound, the hidg_bind() function re-initializes wait queues that still contain registered waiters - corrupting the kernel's internal list structures and triggering an availability-impacting kernel panic upon epoll removal. The vulnerability carries a CVSS score of 5.5 (Medium), EPSS of 0.02% (7th percentile), is not listed in CISA KEV, and no public exploit code has been identified at time of analysis.

Permissive CORS policy in MeTube up to version 2026.04.09 allows remote attackers to bypass same-origin restrictions by exploiting wildcard origin acceptance (cors_allowed_origins='*') in the on_prepare function of app/main.py, potentially enabling cross-domain data theft or unauthorized API access. The vulnerability requires user interaction (UI:P) but grants information disclosure via cross-domain requests. Publicly available exploit code exists; upgrade to 2026.04.10 fixes the issue by restricting origins to a configurable allowlist.

Weak password recovery in D-Link M60 up to version 1.20B02 allows remote attackers to compromise device authentication through manipulation of the /usr/bin/httpd binary, requiring high attack complexity but with publicly disclosed exploit code available. The vulnerability enables information disclosure and potential unauthorized access to device management functions despite the low CVSS score of 2.9 reflecting limited confidentiality impact.

V2Board through version 1.7.4 exposes sensitive server authentication tokens via GET parameters in the UniProxy API endpoint, causing tokens to be recorded in web server access logs, browser history, HTTP Referer headers, and intermediary proxies. An attacker who obtains access to any log source can extract the token and impersonate a proxy server node, potentially intercepting all user traffic passing through that node.

Stack-based out-of-bounds read in Open CASCADE Technology (OCCT) V8_0_0_rc5 VRML parser allows local attackers with low privileges to cause denial of service by submitting a crafted VRML file. The vulnerability stems from unsafe pointer arithmetic in the quoted-string escape handler that reads past a fixed-size stack buffer without bounds validation. CISA SSVC assessment indicates exploitation is not currently active and not readily automatable, suggesting this is a localized attack requiring user interaction with malicious files.

Open CASCADE Technology (OCCT) V8_0_0_rc5 contains multiple out-of-bounds read vulnerabilities and infinite recursion flaws in its IGES and STEP file parsers that can be triggered by maliciously crafted CAD files, resulting in denial of service or memory disclosure. Local attackers with low privilege can exploit these issues without user interaction by supplying crafted IGES or STEP files to applications using OCCT. EPSS-based risk assessment indicates moderate exploitation probability; CISA SSVC framework rates this as non-automated with partial technical impact (denial of service and information disclosure).

Authenticated admin users in V2Board through version 1.7.4 can exploit unsanitized sort parameters in the user management interface to disclose sensitive database information including password hashes and authentication tokens through ORDER BY-based information disclosure. The vulnerability requires admin privileges and does not enable data modification or service disruption, but allows attackers with administrative access to extract confidential user data by analyzing sort order patterns.

Out-of-bounds read in Open CASCADE Technology V8_0_0_rc5 VRML parser allows denial of service when processing crafted VRML files with invalid coordIndex values. The vulnerability exists in VrmlData_IndexedLineSet::TShape geometry processing, where array indices are used without bounds validation, enabling local attackers with user interaction to crash applications through malformed input.

Route Services in Cloud Foundry Routing Release v0.118.0-v0.371.0 and CF Deployment v0.0.2-v54.14.0 allow authenticated malicious developers to bypass application egress rules by configuring route-services that redirect traffic to internal network destinations otherwise unreachable from external networks or the application itself. This affects the scope of the routing infrastructure, enabling information disclosure and potential lateral movement within the Gorouter network.

IBM watsonx.data versions 2.2 through 2.3 fail to enforce proper network segmentation between Kubernetes pods in the Lakehouse component, allowing attackers with network access to the cluster to transfer data between pods without authentication or authorization controls. This integrity vulnerability has a moderate CVSS score of 5.3 and requires adjacent network access and specific configuration conditions to exploit.

IBM watsonx.data intelligence versions 5.2.0, 5.2.1, 5.3.0, and 5.3.1 store user credentials in plain text within local filesystem locations, allowing any local user to read sensitive authentication material without authentication. This information disclosure vulnerability affects confidentiality but not integrity or availability, and requires local filesystem access to exploit.

Format string injection in Notepad++ 8.9.3 Find Results panel handler allows local attackers to cause denial of service and disclose stack memory by distributing malicious nativeLang.xml language pack files that trigger unsafe format string interpretation during search operations. User interaction is required to load the poisoned language pack and perform a search. No active exploitation confirmed, but patch is available from vendor.

Out-of-bounds read in Absolute Secure Access macOS client prior to version 14.50 allows remote attackers with control of a modified server to send malformed packets triggering denial of service. The vulnerability requires high attack complexity (modified server infrastructure and user interaction) and results in availability impact only, with a low CVSS score of 2.3 reflecting limited real-world severity despite network accessibility.

Format string vulnerability in Secure Access client for macOS prior to version 14.50 allows authenticated local attackers to leak sensitive data from process memory via crafted logging input, potentially exposing secrets through log files. The vulnerability requires local access and valid user privileges but no user interaction. Patch is available from the vendor.

GnuTLS with OCSP verification enabled incorrectly accepts revoked server certificates when presented with specially crafted multi-record OCSP responses during TLS handshakes, allowing attackers to bypass certificate revocation checks and establish connections to compromised servers. The vulnerability requires high attack complexity and specific OCSP configuration, affecting Red Hat Enterprise Linux 6-10, Red Hat Hardened Images, and OpenShift Container Platform 4. No public exploit code or active exploitation has been identified at the time of analysis.

Weblate fails to revoke Django REST Framework API tokens when users change their passwords, allowing attackers with knowledge of a user's old credentials to continue accessing APIs indefinitely even after password changes. The vulnerability affects authenticated users on Weblate versions prior to 5.17.1 and requires high attack complexity due to the need for legitimate account access, but carries meaningful risk in multi-user translation environments where password compromise may go undetected.

Keycloak's Account REST API remains partially accessible even when explicitly disabled via the `--features-disabled=account,account-api` flag, allowing authenticated users to read and modify account data through five unprotected endpoints under `/account/v1alpha1/` that lack the required `checkAccountApiEnabled()` access control gate present in four sibling endpoints within the same service class.

Authenticated users with minimal namespace-scoped privileges can obtain administrative credentials for arbitrary OpenShift clusters provisioned through the MCE hub via the assisted-service REST API. The vulnerability exists in AUTH_TYPE=local mode (the only mode available in on-premises deployments), where the local authenticator grants full administrative access to any request bearing a valid JWT with no per-endpoint restrictions. A valid JWT is embedded as plaintext in the InfraEnvStatus.ISODownloadURL, readable by any user with get rights on an InfraEnv object, enabling extraction of kubeadmin passwords and kubeconfigs for all spoke clusters.

Code injection in Qt SVG module allows attackers to execute arbitrary QML/JavaScript when applications load malicious SVG files through Qt Quick's VectorImage component. Exploitation requires local file access and user interaction (opening crafted SVG). While QML execution is more restricted than native code, attackers can still trigger denial of service, exfiltrate application data, or manipulate UI logic depending on the victim application's privilege context. No active exploitation confirmed (not in CISA KEV), but patch available from Qt Project reduces urgency for immediate emergency response.

Authenticated remote attackers can access sensitive personal information in MeWare PDKS versions 16.20200313 through before VMYR_3.5.2025117 due to improper access controls. The vulnerability allows disclosure of private data without authorization, affecting confidentiality. CVSS score of 6.5 reflects moderate severity with network accessibility and low attack complexity, though authentication is required. No active exploitation or public proof-of-concept has been confirmed at time of analysis.

Dancer::Session::Abstract through version 1.3522 generates cryptographically weak session identifiers by combining predictable inputs (file path, process ID, epoch time) with an insufficiently-seeded Perl rand() function, allowing remote attackers to predict valid session IDs and hijack user sessions without authentication. The vulnerability affects Perl-based web applications using Dancer framework's default session handling; active exploitation is not confirmed but the attack requires only guessing a session ID, making it practically exploitable.

Missing CAP_NET_ADMIN capability check in the Linux kernel's rtnetlink subsystem allows a local low-privileged user with an unprivileged user namespace to inject network interfaces into arbitrary network namespaces, including the host's init_net. The flaw is in rtnl_newlink(), which validates capabilities only in the caller's namespace but not in the peer namespace when creating paired devices such as veth, vxcan, and netkit. No active exploitation is confirmed (not in CISA KEV), and the EPSS score of 0.02% (4th percentile) reflects very low current exploitation probability, though the impact is rated Availability:High and the vulnerability is relevant to multi-tenant and containerized environments.

Apache Airflow's SmtpHook performs STARTTLS upgrades without SSL certificate validation, allowing man-in-the-middle attackers to intercept SMTP credentials. Remote unauthenticated attackers positioned between an Airflow worker and SMTP server can present a self-signed certificate, complete the TLS handshake, and capture login credentials sent after the upgrade. The vulnerability affects apache-airflow-providers-smtp versions 2.0.0 through 2.x and is patched in version 3.0.0 or later. No public exploit code identified at time of analysis, but EPSS score of 0.01% indicates low real-world exploitation probability despite confidentiality impact.

Out-of-bounds read in Exim before 4.99.2 when UTF-8 operators are enabled allows remote unauthenticated attackers to leak sensitive information through error messages by sending email with malformed UTF-8 trailing characters in headers. The vulnerability has high attack complexity due to the requirement for UTF-8 operator enablement and specific malformed input crafting, but requires no user interaction and operates over the network on default deployments.

Insecure default file permissions in the Anthropic TypeScript SDK's `BetaLocalFilesystemMemoryTool` expose persisted agent state to local read access and, in Docker deployments with permissive umasks, enable local users to overwrite memory files and manipulate subsequent model behavior. Affected versions span `@anthropic-ai/sdk` 0.79.0 through 0.91.0; a fix is available in 0.91.1. No public exploit exists and exploitation has not been observed (SSVC: none), with an EPSS score of 0.01% placing this in the 2nd percentile - risk is real but narrowly scoped to shared-host or containerized deployments.

Admidio members_assignment_data.php endpoint leaks hidden profile field values through a blind search oracle attack. Role leaders with ROLE_LEADER_MEMBERS_ASSIGN permissions can infer exact values of hidden PII fields (birthdays, street addresses, cities, postal codes, countries) by observing which users appear in search results, despite these fields being suppressed from JSON output. The vulnerability affects Admidio versions up to 5.0.8 and is fixed in 5.0.9.

CKAN fails to validate SMTP server certificates, allowing attackers to spoof the configured mail server with any certificate including self-signed ones and intercept SMTP credentials and email content via man-in-the-middle attack. Versions below 2.10.10 and 2.11.0 through 2.11.4 are affected. Vendor-released patches are available in CKAN 2.10.10 and 2.11.5.

GNU wget2 incorrectly validates TLS server certificates, accepting certificates with improper Key Usage (KU) or Extended Key Usage (EKU) attributes. This allows attackers who have compromised certificates issued for non-server purposes to impersonate legitimate servers in TLS connections, enabling man-in-the-middle attacks that leak sensitive information such as authentication credentials or request/response data.

Denial of service via crafted ICMPv6 Router Advertisement in FreeRTOS-Plus-TCP affects embedded devices running versions 4.0.0-4.2.5 and 4.3.0-4.4.0, crashing the device when a truncated PREFIX_INFORMATION option triggers an out-of-bounds read in the IPv6 RA parser. An adjacent-network attacker with no privileges or user interaction can repeatedly crash affected IoT/embedded devices, causing persistent availability loss. No active exploitation has been confirmed (not in CISA KEV) and EPSS sits at 0.01% (3rd percentile), but vendor-confirmed patches exist for both affected branches.

Arbitrary file write via symlink following in Dell/Alienware Purchased Apps (versions prior to 1.1.31.0) allows a local low-privileged attacker to redirect application file operations to unintended targets, potentially overwriting sensitive system files and causing integrity or availability damage. The high attack complexity (AC:H) indicates a race condition or precise timing requirement, limiting opportunistic exploitation. No active exploitation has been identified - SSVC assessment rates exploitation as none and the vulnerability is absent from CISA KEV - though successful exploitation could yield significant local impact.

Content-type validation bypass in AgentFlow's localhost web API allows browser-driven cross-origin requests to reach the POST /api/runs and POST /api/runs/validate control plane endpoints without enforcement of application/json. Affected versions prior to commit 1667fa3 (CPE: cpe:2.3:a:berabuddies:agentflow:*) fail to reject non-JSON content types, enabling a malicious web page visited by a local AgentFlow user to issue CORS-simple requests that bypass the browser's same-origin enforcement and submit pipeline operations to the local agent runtime. No active exploitation is confirmed in CISA KEV; the SSVC framework records a PoC with partial technical impact and non-automatable exploitation.

Symlink following in NousResearch hermes-agent 0.8.0 allows local low-privileged users to bypass sensitive filesystem path protections and write to restricted system files. The `_check_sensitive_path` function in `tools/file_tools.py` validated only the fully-resolved real path, not the original expanded path, enabling bypass on platforms where system directories are themselves symlinks (e.g., macOS `/etc` → `/private/etc`). A publicly available PoC exists; no confirmed active exploitation (not in CISA KEV), and the CVSS 4.0 score of 1.9 with EPSS at the 2nd percentile indicates low real-world risk.

MongoDB Server's user management command contains an authorization flaw allowing authenticated low-privilege users to make limited modifications to authentication-related data belonging to other user accounts. Affected across three active release branches (7.0.x, 8.0.x, 8.2.x), this could allow one MongoDB user to interfere with how authentication is processed for another account - a form of limited horizontal privilege escalation. No public exploit exists and SSVC confirms no observed exploitation, but the network-accessible attack vector and low authentication requirement warrant timely patching in multi-tenant or shared MongoDB environments.

Local denial-of-service in Netskope Client's Endpoint DLP Module on Windows allows an unprivileged local user to crash the operating system via a kernel-level out-of-bounds read in a driver, producing a Blue-Screen-of-Death. Exploitation requires the Endpoint DLP module to be active in the client configuration, limiting the exposed population to deployments where that feature is intentionally enabled. No public exploit code exists and no active exploitation has been identified; EPSS is 0.01% and SSVC rates exploitation as none, consistent with the local-only attack vector.

Unsafe deserialization in Jenkins Matrix Authorization Strategy Plugin versions 2.0-beta-1 through 3.2.9 allows an authenticated user with Item/Configure permission to instantiate arbitrary Java classes from the Jenkins classpath by specifying class names in job configuration. The vulnerability arises because the plugin invokes parameterless constructors of user-specified classes when deserializing inheritance strategies without type restriction. No public exploit is identified at time of analysis, and SSVC signals exploitation as none, but the real-world impact depends on the richness of the Jenkins classpath - gadget classes available in that environment determine the attacker's effective reach.

Sensitive system information in the BetterDocs WordPress knowledge base plugin (WPDeveloper, all versions through 4.3.10) is exposed to unauthenticated remote attackers via CWE-497, enabling retrieval of embedded sensitive data without credentials. Reported by Patchstack and tracked as EUVD-2026-26215, this medium-severity finding carries an EPSS of 0.03% (8th percentile) with no public exploit code or active exploitation recorded at time of analysis. The SSVC framework confirms no known exploitation but flags the attack as automatable, making it a viable target for mass WordPress plugin scanning campaigns.

PackageManagerService on Samsung Mobile Devices running Android 14, 15, and 16 allows local attackers without elevated privileges to bypass application installation restrictions by exploiting insufficient verification of data authenticity. The affected component fails to adequately validate the integrity of data governing app installation policies, enabling unauthorized modification of those restrictions on unpatched devices. No public exploit code has been identified and SSVC rates exploitation as none, but the impact is significant in enterprise MDM and parental-control deployments where installation restrictions are an active security control.

Cross-origin data leakage in Google Chrome's MHTML implementation exposes sensitive data from other browser origins to a malicious extension. Versions prior to 147.0.7727.138 are affected; exploitation requires convincing the target user to install an attacker-controlled Chrome Extension that triggers a race condition in MHTML processing. No public exploit code has been identified and CISA SSVC confirms no exploitation at time of analysis, placing real-world risk significantly below what the product's ubiquity might suggest.

Fiber v3 cache middleware's default KeyGenerator discards the query string entirely, keying responses solely on the request path, which causes requests to distinct URIs such as /?id=1 and /?id=2 to share a single cache slot. Any network-accessible application using default configuration on query-variant endpoints will serve the first cached response to all subsequent callers regardless of their query parameters, creating a cache poisoning-like information disclosure and integrity corruption scenario. A working proof-of-concept is embedded in the GHSA advisory and an existing test case in the repository confirms the behavior; SSVC rates exploitation as automatable, though EPSS remains very low at 0.03% with no CISA KEV listing, indicating no observed mass exploitation as of analysis time.

Hard-coded cryptographic key exposure in Xuxueli xxl-job's OpenAPI endpoint (versions 3.3.0-3.3.2) allows remote attackers to authenticate to the admin scheduling API using a known default token (`default_token`) without supplying legitimate credentials. A proof-of-concept exploit has been publicly disclosed via the upstream GitHub issue tracker (SSVC: poc), though CVSS AC:H and SSVC's non-automatable rating indicate mass exploitation is constrained - only deployments retaining the default token configuration are exposed. No confirmed active exploitation (not in CISA KEV) and a low EPSS of 0.04% (14th percentile) suggest targeted rather than opportunistic risk.

Improper resource identifier control in Xuxueli xxl-job versions 3.3.0 through 3.3.2 exposes execution log entries to unauthorized disclosure via logId parameter manipulation in the admin Execution Log Handler. The commit diff reveals that `JobGroupPermissionUtil.validJobGroupPermission` was commented out in `JobLogController.java`, effectively removing the job-group-level access boundary and enabling IDOR-style cross-boundary log reads. A public proof-of-concept exploit exists per SSVC (Exploitation: poc), though a CVSS 4.0 score of 2.9 and EPSS of 0.05% (16th percentile) reflect that high attack complexity and limited confidentiality impact constrain real-world risk; no active exploitation is confirmed in CISA KEV.

Cryptographic signature verification bypass in Elastic Package Registry versions 0.1.0 through 1.37.0 allows an adversary-in-the-middle, or an attacker with influence over a self-hosted registry's served content, to substitute tampered packages that are accepted without triggering integrity check failures. The flaw (CWE-347) undermines supply-chain integrity for organizations running self-hosted registry instances. No active exploitation is confirmed (SSVC: exploitation none; EPSS 0.02%), but the high integrity impact and zero-privilege-required vector make patching to 1.38.0 a priority for self-hosted deployments.

Cross-channel local file exfiltration in OpenClaw (npm package, all versions before 2026.4.8) allows an attacker with low-privilege access to craft malicious shared reply MEDIA path references that cause a separate channel to resolve and expose local file contents as trusted generated media. The root cause (CWE-73) is improper neutralization of path references in shared reply MEDIA URIs: OpenClaw accepts these references without enforcing a trust origin check, effectively treating attacker-controlled paths as internally generated content. No public exploit code has been identified and CISA has not listed this in KEV, but the high confidentiality impact on the vulnerable system (VC:H in the CVSS 4.0 vector) and the arbitrary local-file-read capability make patching a priority for any affected deployment.

Git environment variable injection in OpenClaw (npm) before 2026.4.8 allows a local low-privileged attacker to redirect git operations by setting GIT_DIR and related plumbing variables in the process environment prior to host exec calls. The root cause is an incomplete denylist (CWE-184) that strips some but not all dangerous environment variables before shelling out. No public exploit or active exploitation has been identified at time of analysis; EPSS of 0.01% (3rd percentile) and SSVC exploitation status of 'none' corroborate low near-term exploitation likelihood.

OpenClaw before version 2026.4.2 leaks shared-secret length information through timing side-channel attacks in cryptographic comparison operations. The vulnerability stems from early length-mismatch checks in shared-secret comparison routines that violate constant-time security requirements, allowing remote attackers to measure timing differences and infer secret lengths without authentication. This weakens the cryptographic guarantees of the library's shared-secret handling.

OpenClaw before version 2026.3.31 accepts arbitrary tailnet peers as DNS authorities due to improper validation in its wide-area discovery mechanism, enabling attackers positioned within the same tailnet with CA-trusted endpoint access to manipulate DNS resolution and exfiltrate operator credentials. The vulnerability requires adjacent network access, high attack complexity, and user interaction, but results in high confidentiality impact through credential theft. No active exploitation has been publicly confirmed at the time of analysis.

OpenClaw before version 2026.3.31 fails to block plugin installation when security scans detect threats, allowing authenticated users to install malicious plugins by ignoring visible scan warnings. The vulnerability requires user interaction (UI:P) and authenticated access (PR:L), but enables installation of untrusted code with moderate integrity impact when exploited.

Server-side request forgery in NVIDIA NemoClaw's validateEndpointUrl() function allows local attackers with user interaction to supply crafted endpoint URLs targeting the 0.0.0.0/8 address range via blueprint configuration files or CLI flags, leading to information disclosure. The vulnerability affects all versions of NemoClaw and requires local access with user interaction to trigger, limiting exposure to systems where untrusted users can modify configuration or invoke CLI commands.

NVIDIA Flare SDK is vulnerable to path traversal via improper input validation, allowing authenticated remote attackers to disclose sensitive information. The vulnerability affects all versions of the SDK and requires valid user credentials to exploit, making it a moderate-risk issue for organizations using Flare in multi-user environments. No public exploit code or active exploitation has been identified.

Account lockout abuse in Ercom Cryptobox allows any authenticated user to deny another user access by sending a specially crafted request that triggers the platform's lockout mechanism. Specific versions in the 4.37-4.40 range are confirmed affected per ENISA EUVD-2026-26070, and the vendor has released a patch. No active exploitation is confirmed - EPSS stands at 0.04% (13th percentile) and CISA SSVC marks exploitation status as none, placing this firmly in the category of a real but low-urgency availability risk.

Spring gRPC versions 1.0.0 through 1.0.2 leak sensitive authentication failure details in gRPC status descriptions to unauthenticated remote callers, enabling reconnaissance for follow-up attacks. The vulnerability exposes raw server-side AuthenticationException messages without sanitization, providing attackers with information about authentication mechanisms and potential weaknesses. This low-severity information disclosure (CVSS 3.7) requires high attack complexity but affects default configurations.

Unauthenticated remote attackers can access sensitive documents containing personally identifiable information (PII) in Aranda Service Desk versions prior to 8.3.12 by exploiting predictable log file names in the Aranda File Server (AFS) component. Attackers retrieve daily activity logs from a publicly accessible directory to obtain direct virtual paths of uploaded files, then bypass access controls to download the documents. CISA SSVC framework confirms proof-of-concept code exists and the vulnerability is fully automatable, significantly lowering the barrier to exploitation despite no confirmed active exploitation at time of analysis.

GNU nano creates the ~/.local directory with world-writable permissions (0777) on first use of XDG data storage features when the directory does not exist, allowing local attackers in systems with relaxed umasks (such as containers, CI/CD runners, or environments with umask 000) to write attacker-controlled files into the victim's XDG directory hierarchy via a race condition. The vulnerability affects nano versions before 9.0 and carries a CVSS score of 2.1 with CISA SSVC assessment indicating partial technical impact but no known public exploitation.

OpenShift Container Platform build system allows authenticated users with the edit ClusterRole to inject arbitrary environment variables into docker-build containers via the buildconfigs/instantiate API, enabling information disclosure attacks such as build traffic interception through LD_PRELOAD or http_proxy manipulation. This represents an incomplete remediation of a prior vulnerability, affecting confidentiality of sensitive build data with CVSS 4.3 (network-accessible, low complexity, authenticated). No public exploit code or active exploitation has been confirmed at the time of analysis.

Remote command execution in mpGabinet 23.12.19 and below allows authenticated database administrators or unauthenticated attackers (via chained exploitation of CVE-2026-40550 and CVE-2026-40551) to achieve system command execution by manipulating attachment storage paths in the database to reference attacker-controlled resources that execute when users open the files. The vulnerability requires direct database access and user interaction to trigger execution, but becomes unauthenticated when chained with companion CVE vulnerabilities that grant database and application access.

Spring AI versions 1.0.0-1.0.5 and 1.1.0-1.1.4 expose ONNX machine learning models to unauthorized disclosure when the application runs in shared hosting environments, allowing local users with limited system access to read sensitive model files and potentially reverse-engineer proprietary ML logic. The vulnerability stems from insecure temporary file handling (CWE-377) that fails to restrict file permissions on extracted model artifacts. Authentication requirements are minimal-only local system access is needed-making this a significant risk in multi-tenant cloud platforms and shared servers.

Penetration Testing engineers at Amazon have discovered a flaw where the camera system fails to properly handle data supplied in certain requests, causing a service disruption. Rated medium severity (CVSS 5.3), this vulnerability is remotely exploitable, low attack complexity.

Out-of-bounds read in Artifex MuPDF up to version 1.28.0 within the CFF Index Handler's fz_subset_cff_for_gids function allows local attackers with low privileges to disclose sensitive information from application memory. The vulnerability requires local access and low privilege level but can be triggered without user interaction; publicly available exploit code exists and the vulnerability remains unpatched as of the last vendor response.

Out-of-bounds read in Apache Thrift C++ JSON deserialization allows remote attackers to leak sensitive information and trigger denial of service via malformed JSON payloads. Affects Apache Thrift versions prior to 0.23.0. The vulnerability has low exploitation probability (EPSS 0.02%) and is not currently listed in CISA KEV, suggesting limited real-world weaponization despite network-accessible attack vector.

OpenClaw before version 2026.3.31 allows authenticated attackers to read arbitrary host files through improper validation in the appendLocalMediaParentRoots function, enabling exfiltration of credentials and sensitive data. The vulnerability permits model-initiated file access by exploiting a self-whitelisting mechanism that fails to properly validate media parent directory paths. Authentication is required, but the flaw affects confidentiality with a CVSS score of 6.0.

OpenClaw versions 2026.2.19 before 2026.3.31 allow authenticated attackers to suppress legitimate webhook events across different accounts in multi-tenant deployments by exploiting improper cache isolation in the Zalo webhook replay-deduplication mechanism. An attacker with control of one authenticated Zalo webhook path can match event_name and message_id parameters to suppress events on victim accounts, causing denial of service to webhook processing. No public exploit code or active exploitation has been identified, though the vulnerability requires valid authentication and incremental exploitation (AT:P), limiting immediate risk.

Spring Boot applications configured with ApplicationPidFileWriter are vulnerable to local file corruption when a high-privilege user can write to the PID file directory. An attacker with high privileges and write access to the PID file location can corrupt arbitrary files each time the application restarts, achieving denial of service or data integrity violations. Exploitation requires local access and elevated privileges, limiting real-world impact to co-resident or insider threat scenarios. No active exploitation has been publicly reported.

KDE Dolphin before 25.12.3 allows sandboxed applications (running under Flatpak or AppArmor confinement) to bypass sandbox restrictions and open arbitrary files outside their containment boundary through the FileManager1 D-Bus protocol implementation. An attacker controlling a sandboxed application can exploit this to access sensitive files or execute scripts with user interaction, circumventing the intended isolation model.

Uncaught exception in UERANSIM up to v3.2.7 allows remote unauthenticated attackers to crash the Radio Link Simulation (RLS) message decoder by sending a crafted PDU with an oversized or malformed pduLength field, causing denial of service against 5G simulation sessions. The SSVC framework marks this as automatable with a POC exploit available, though real-world impact is constrained to research and testing environments given the product's deployment profile. A vendor patch (v3.2.8) was released promptly after responsible disclosure, confirmed by the upstream GitHub release tag.

Spring Boot's Elasticsearch auto-configuration omits hostname verification when an SSL bundle is used, opening TLS connections to the Elasticsearch server to man-in-the-middle interception by an attacker on the adjacent network. Affected versions are 4.0.0 through 4.0.5; only deployments explicitly configured with an SSL bundle are exposed. No active exploitation is confirmed (absent from CISA KEV, EPSS 0.01% at the 3rd percentile), and VMware has released a fix in version 4.0.6.

Unauthenticated directory traversal in Pro-Bit versions before 1.77.4 exposes sensitive directories and subdirectories to remote attackers without authentication. The vulnerability allows direct access to protected file system locations via network requests, enabling unauthorized information disclosure. EPSS score of 0.02% (6th percentile) indicates low observed exploitation probability in the wild, and no CISA KEV listing exists at time of analysis, suggesting limited active exploitation despite the CVSS 7.5 severity rating.

Elevation of Privileges in Dell Alienware Command Center (AWCC) versions prior to 6.13.8.0 allows a local low-privileged attacker to operate beyond their intended permission level due to a CWE-272 Least Privilege Violation. The CVSS vector (AV:L/AC:H/PR:L) confirms this is a local attack requiring an existing low-privileged foothold and elevated complexity conditions, with primary impact on system integrity (I:H). No public exploit or active exploitation has been identified at time of analysis; SSVC rates exploitation as 'none' and technical impact as 'partial', consistent with the moderate 5.3 base score.

Indefinite deadlock in the Linux kernel's igb Gigabit Ethernet driver causes permanent TX queue stall when an AF_XDP zero-copy application is killed abruptly. Affected systems using Intel igb-family NICs (82575, 82576, I350) experience a hang in igb_down() because napi_synchronize() waits for NAPI_STATE_SCHED to clear - a state that igb_clean_rx_irq_zc() prevents from clearing by continuously returning the full NAPI budget after the XSK buffer pool is destroyed. The TX watchdog fires and the network interface becomes permanently unavailable until reboot. No public exploit has been identified (EPSS 0.02%, 4th percentile), and upstream patches are available in stable kernel branches 6.18.23, 6.19.13, and 7.0.

Denial of service in Linux kernel GPIO OMAP driver allows local authenticated users to crash the system via a deadlock condition triggered by improper driver registration from probe() callback. The vulnerability stems from registering the omap_mpuio_driver within omap_gpio_probe(), which violates driver core locking rules and creates a potential deadlock when device_lock enforcement was strengthened in commit dc23806a7c47. EPSS score of 0.03% reflects low exploitation probability despite availability of patched kernel versions.

Improper authorization in Wooey versions up to 0.13.2 permits any authenticated non-staff user to upload or update Python scripts via the add_or_update_script API endpoint in wooey/api/scripts.py, bypassing the intended staff-only privilege boundary. Because Wooey executes uploaded scripts on the underlying server, this flaw carries escalation potential beyond the limited CVSS 4.0 base score of 2.1 in multi-tenant or internet-facing deployments. A proof-of-concept has been publicly disclosed; the vulnerability is not in CISA KEV, and EPSS is negligible at 0.04% (13th percentile), suggesting exploitation remains opportunistic rather than widespread.

Stale KV cache data leakage in vllm's KV Block Handler (versions 0.1-0.19.0) allows partial disclosure of prior inference requests' attention key/value data to subsequent requests. The `needs_kv_cache_zeroing` property incorrectly returned `False` for FullAttention model configurations - relying solely on `has_mamba_layers` - causing recycled KV cache blocks to go unzeroed between requests; partial-block tail slots retaining NaN/Inf values from prior sessions could then leak into the masked softmax computation of the next request. A publicly available PoC exists per SSVC and CVSS 4.0 E:P, though EPSS at 0.05% (14th percentile) and absence from CISA KEV indicate very low observed exploitation in practice.

Transient execution side-channel exposure in AMD EPYC 7001 Series and EPYC Embedded 3000 Series processors permits a local low-privileged attacker to leak limited sensitive data through the floating point divisor unit's microarchitectural state during speculative execution. Exploitation is constrained to local access with specific attack conditions present (CVSS 4.0 AT:P), placing this firmly in the Spectre/Meltdown class of hardware-level vulnerabilities. No public exploit code or active exploitation has been identified; EPSS sits at 0.01% (1st percentile) and CISA SSVC rates exploitation as none with non-automatable attack path.

Out-of-bounds read in GPAC's elng_box_read function allows local attackers to trigger partial information disclosure and potential availability impact by supplying a crafted MP4 file with an oversized Extended Language Tag (elng) box. The fix commit confirms the root cause: ptr->size is not validated against GF_UINT_MAX before use, permitting a read beyond allocated memory. A publicly available proof-of-concept exploit exists (GitHub issue #3516), though EPSS sits at 0.01% and SSVC classifies exploitation as non-automatable with only partial technical impact - placing real-world priority well below the noise floor for most organizations.

Module import confusion in pip before version 26.1 allows a maliciously crafted wheel package to execute arbitrary code within pip's own process by exploiting the sequencing of pip's self-update check relative to wheel installation. Because pip deferred well-known module imports for startup performance and triggered the self-update check after wheels were installed, a freshly installed malicious module could shadow a legitimate import targeted by the self-update routine. No public exploit code is identified at time of analysis, EPSS sits at the 2nd percentile, and SSVC confirms no exploitation, though the CWE-829 root cause carries inherent supply-chain risk.

Improper certificate validation in Apache Storm Prometheus Reporter versions 2.6.3 to 2.8.6 allows man-in-the-middle attacks across all TLS connections in the Storm daemon when the skip_tls_validation configuration option is enabled. Enabling this setting for Prometheus PushGateway connections inadvertently downgrades the JVM-wide SSL context, causing all subsequent HTTPS communications (ZooKeeper, Thrift, Netty, UI) to trust arbitrary certificates without validation, enabling interception of cluster state, topology submissions, and administrative credentials. No public exploit code identified at time of analysis, and EPSS scoring of 0.01% reflects the requirement for explicit administrator misconfiguration to trigger the vulnerability.

Unhandled exception in Foxit PDF Editor and Foxit PDF Reader allows local denial of service when a user opens a maliciously crafted PDF file with insufficient parameter verification, causing the application to crash via an uncaught std::invalid_argument exception. The vulnerability requires user interaction (opening a file) and local file access but affects all versions of both products without authentication requirements.

Weak cryptographic hash usage in code-projects Chat System 1.0 allows remote attackers to compromise password security through the MD5 Hash Handler in update_user.php. The vulnerability stems from use of MD5 for password hashing, a cryptographically broken algorithm that enables rapid offline cracking of password hashes. Publicly disclosed exploit code exists, though exploitation requires high attack complexity. The vulnerability impacts password confidentiality with low direct severity but creates substantial downstream risks for user account compromise.

OPPO Wallet APP contains a trusted domain validation bypass flaw that allows local attackers with user interaction to hijack account tokens and disclose sensitive information. The vulnerability affects all versions of OPPO Wallet APP and exploits improper domain verification in protected interface access controls, enabling token theft through local attack vector.

uriparser before 1.0.1 suffers a numeric truncation vulnerability in text range comparison that causes denial of service when processing URIs with gigabyte-scale lengths. The flaw occurs because internal range comparisons truncate large numeric values, allowing maliciously crafted oversized URIs to bypass length validation and trigger memory exhaustion or processing failures. Local attackers can exploit this via specially constructed input, though practical exploitation requires an application to accept and process URIs of exceptional size.

Improper ownership management in Moxa Secure Router allows low-privileged authenticated users to access exported configuration files containing hashed administrative passwords, enabling credential disclosure. The vulnerability is confined to scenarios where configuration files have been exported and requires valid user credentials to exploit; no impact to system integrity or availability has been identified.

CodeAstro Online Job Portal 1.0 exposes file and directory information through the /users/user-cvs/ endpoint via remote unauthenticated access, allowing attackers to enumerate and retrieve sensitive resume and user data. The vulnerability has publicly available exploit code and affects all versions of the application via the CPE cpe:2.3:a:codeastro:online_job_portal:*:*:*:*:*:*:*:*. CVSS 5.5 with confirmed public exploit availability and EPSS exploitation probability indicates moderate real-world risk for deployments accessible over the network.