Information Disclosure

Apache Thrift versions prior to 0.23.0 are vulnerable to a denial-of-service condition with unspecified attack mechanisms related to CWE-789 (uncontrolled memory allocation). The vulnerability affects multiple language implementations including Rust, Java, and Node.js, and can be triggered remotely without authentication or user interaction, though the technical mechanism remains partially obscured in available disclosures. With EPSS score of 0.02% (percentile 5%), active exploitation appears unlikely despite the low CVSS complexity score.

TLS hostname verification is disabled in Apache Thrift's Java TSSLTransportFactory implementation (versions prior to 0.23.0), allowing remote unauthenticated attackers to perform man-in-the-middle attacks against encrypted communications. The vulnerability enables interception and potential modification of data in transit with low attack complexity and no user interaction required. While EPSS shows minimal current exploitation activity (0.00%), CISA SSVC classifies this as automatable with partial technical impact, and a vendor patch is available in version 0.23.0.

Arbitrary kernel memory read/write in Realtek rtl819x Jungle SDK Wi-Fi driver allows local unprivileged attackers to access and modify kernel memory through debug ioctl handlers (0x89F5/0x89F6) that were left enabled in production builds. All known SDK versions through v3.4.14B are affected. A publicly available exploit exists (GitHub repository CVE-2026-36355), enabling privilege escalation and data exfiltration on devices using the vulnerable rtl8192cd driver. EPSS data unavailable; not currently in CISA KEV.

OpenStack Horizon 25.6 and 25.7 before 25.7.3 allows unauthenticated remote attackers to exhaust session storage backend resources through repeated requests that trigger write operations prior to authentication, causing denial of service. This is a regression of CVE-2014-8124 and is assigned CVSS 5.3 (network-based, low complexity, no authentication required).

Credential forwarding vulnerability in OpenStack Ironic's idrac driver allows authenticated attackers to steal time-limited Keystone tokens or molds storage credentials by manipulating import operations. Attackers with low-privileged Ironic access can redirect these credentials to attacker-controlled endpoints, gaining unauthorized access to all OpenStack services that Ironic is authorized for. Fixed in versions 26.1.6, 29.0.5, 32.0.1, and 35.0.1. CVSS 7.7 with scope change (S:C) reflects the privilege escalation from Ironic-only access to full OpenStack service access.

Authenticated users with non-admin SETTINGS permission in pyload-ng ≤0.5.0b3.dev99 can redirect all outbound HTTP traffic through attacker-controlled proxies by modifying ungated proxy configuration fields (enabled, host, port, type). The vulnerability is an incomplete fix in a series of authorization bypass issues (CVE-2026-33509/-35463/-35464/-35586) where a hand-maintained allowlist in set_config_value() gates only proxy credentials (username/password) but not the proxy destination itself, allowing credential theft, traffic interception, and response injection across downloads, captcha solvers, and update checks. No public exploit identified at time of analysis, though the GitHub advisory includes a working proof-of-concept demonstrating traffic redirection via API calls. Patch confirmed in version 0.5.0b3.dev100 per vendor advisory GHSA-pg67-9wjv-mr85.

Man-in-the-middle attackers can strip TLS protection from Ruby net-imap STARTTLS connections by injecting a premature tagged OK response with a predictable tag. The vulnerability allows attackers to bypass TLS encryption, forcing the client to transmit credentials and email content in cleartext while the application believes the connection is secure. Vendor-released patches (net-imap 0.6.4, 0.5.14, 0.4.24, 0.3.10) are available. CVSS 7.6 severity reflects network-accessible attack with low complexity but requires man-in-the-middle positioning. No public exploit code identified at time of analysis, though the attack mechanism is well-documented in security research (NO STARTTLS project).

Package substitution in Chainguard apko allows network-positioned attackers to inject arbitrary APK packages into container images during build. apko verifies APKINDEX.tar.gz signatures but fails to validate individual downloaded .apk packages against signed index checksums, accepting mismatched packages silently. Attackers controlling download responses (compromised mirrors, HTTP repositories, poisoned CDN caches) can replace legitimate packages with malicious ones. Fixed in version 1.2.7. CVSS 7.5 with network vector and no authentication required. EPSS data not available; no CISA KEV listing indicates targeted rather than widespread exploitation at time of analysis.

Authenticated attackers can escalate privileges to administrator in Pelican Web User Interface versions 7.21 through 7.24 by manipulating database records before legitimate admin users log in. This vulnerability was discovered by a Claude coding agent on April 2, 2026, and affects servers with Server.UIAdminUsers or Server.AdminGroups configured where designated admins have not previously authenticated. No public exploit code exists, and Pelican Command Line reports no confirmed exploitation in OSDF-managed services. Vendor patches are available across all affected minor release series (>=v7.21.5, >=v7.22.3, >=v7.23.3, >=v7.24.2), with fix commit 7f73b9c3e677 addressing CWE-863 (Incorrect Authorization).

Quarkus OpenAPI Generator versions before 2.16.0-lts and 2.16.0 (fixed in 2.17.0) send authentication credentials to unintended API endpoints due to overly broad path-parameter regex matching. The generated authentication filter treats OpenAPI path-template placeholders like {param} as .* patterns, allowing a single parameter to consume forward slashes and match multiple distinct operations. This causes bearer tokens, OAuth tokens, API keys, and basic credentials configured for one protected operation to be leaked to different, unprotected operations on the same service when a client invokes them through normal generated-code paths. No public exploit code has been identified, but the vulnerability is trivial to trigger and affects all authentication schemes relying on the shared path-matching logic.

Deactivated user accounts in CI4MS retain full backend access until session expiration because the authentication filter fails to re-check the active status field for existing sessions. When an administrator sets a user's active field to 0, the user's session cookie remains valid and auth()->loggedIn() continues returning true, allowing the deactivated user to access protected resources for up to 7200 seconds (default session timeout). The vulnerability exists because code to enforce this check was commented out since the filter's initial commit and never activated.

Arbitrary code execution on RPM packagers' workstations occurs when pyp2spec generates spec files from malicious PyPI package metadata containing unescaped RPM macro directives. Affects pyp2spec versions prior to 0.14.1, primarily impacting Fedora/RHEL packagers who process untrusted PyPI packages. Attack triggers during spec file parsing (rpmbuild -bs, rpm -q --specfile) before any build step, enabling compromise via typosquatted packages or those under Fedora review. Exploited packager workstations hold dist-git SSH keys, Koji build credentials, and Bodhi update rights, creating supply chain risk. Vendor-released patch 0.14.1 available per GitHub advisory GHSA-r35x-v8p8-xvhw. No public exploit identified at time of analysis, but attack complexity is low with realistic targeting scenarios described.

Argo Workflows executor logs artifact repository credentials in plaintext to pod logs during artifact operations, exposing S3 access/secret keys, GCS service account keys, Azure storage keys, and Git passwords. Users with Kubernetes RBAC permissions to read pod logs in the workflow namespace can extract these credentials directly from workflow execution logs. This vulnerability affects Argo Workflows v4.0.0 through v4.0.4 and represents an incomplete fix of CVE-2025-62157. Vendor-released patch (v4.0.5) is available with GitHub commit bdd40908 removing credential-bearing struct logging. No public exploit identified at time of analysis, though exploitation is trivial given the included working proof-of-concept YAML.

nginx-ui versions prior to 2.3.8 expose 40+ protected configuration fields through the GetSettings API to authenticated users, including JwtSecret (auth token forgery), NodeSecret (cluster impersonation), and OIDC ClientSecret (OAuth takeover). The protected tag is enforced only during writes but completely ignored during reads, allowing authenticated attackers to extract sensitive secrets and IP whitelist configurations without requiring additional privileges or user interaction.

nginx-ui prior to version 2.3.8 exposes sensitive configuration values including node.secret via an authenticated GET /api/settings endpoint, allowing an authenticated user to retrieve the shared authentication secret and subsequently impersonate the init administrative user by sending requests with the stolen node.secret via the X-Node-Secret header or node_secret query parameter. This enables privilege escalation and full administrative access to the Nginx configuration interface without additional authentication.

Missing authorization checks in Argo Workflows v4.0.0-4.0.4 allow any authenticated user-even those with fake Bearer tokens-to create, read, update, and delete Kubernetes ConfigMaps containing workflow synchronization limits. The ConfigMap-backed sync provider (server/sync/sync_cm.go) completely omits auth.CanI permission validation on all four CRUD endpoints. Publicly available exploit code exists (detailed PoC in advisory). CVSS 8.5 reflects network-accessible authentication bypass enabling high integrity/availability impact through denial-of-service and arbitrary ConfigMap manipulation. Patch released in version 4.0.5 adding checkConfigMapPermission() calls to validate Kubernetes RBAC before operations.

Traefik's errors middleware discloses sensitive HTTP headers including Authorization and Cookie to separate error page services when backends return configured error status codes. Affected versions are Traefik v2.11.43 and earlier, v3.6.14 and earlier, and v3.7.0-rc.0 through v3.7.0-rc.2. The vulnerability allows credentials meant only for backend services to be forwarded to distinct error page infrastructure, expanding exposure across service boundaries. Vendor-released patches are available; actively exploited status not confirmed.

Remote unauthenticated attackers can bypass ExifTool tag blocklist in Gotenberg 8.x via group-prefixed tag names (e.g., 'System:FileName' instead of 'FileName'), enabling arbitrary file renaming, relocation, and permission modification within the container filesystem. One HTTP request exploits this input validation bypass (CWE-20) to circumvent protections from a prior security fix (GHSA-qmwh-9m9c-h36m). The vulnerability affects all metadata-accepting endpoints in Gotenberg's default configuration, which typically runs without authentication. No public exploit code is confirmed, but a detailed proof-of-concept is published in the GitHub advisory (GHSA-62p3-hvxx-fxg4). CVSS 8.2 reflects network vector with no authentication required, though real-world impact depends on container isolation and shared volume configurations.

Denial of service in Incus prior to version 7.0.0 allows authenticated users to crash the Incus daemon by importing a maliciously crafted backup archive with physical snapshot directories but tampered metadata arrays. The vulnerability stems from an incorrect bounds check (len(slice) >= i-1 instead of len(slice) > i) in the backup restore and migration code paths, enabling out-of-bounds array access that triggers a runtime panic. Repeated exploitation keeps the Incus service offline, confirmed by a publicly available proof-of-concept.

Prometheus monitoring system exposes Azure AD OAuth client secrets in plaintext via its /-/config HTTP API endpoint. Versions prior to 3.5.3 and 3.11.3 incorrectly type the client_secret field as a plain string instead of Prometheus's redacted Secret type, allowing remote unauthenticated attackers to retrieve sensitive Azure credentials from any exposed Prometheus instance configured for Azure AD remote write. The vulnerability has low exploitation complexity (CVSS AV:N/AC:L/PR:N) with 7.5 severity. Vendor-confirmed patches available in versions 3.5.3 and 3.11.3 (GitHub releases confirmed). EPSS data not provided; no CISA KEV listing indicating targeted exploitation campaigns at time of analysis.

Unauthenticated attackers can enumerate and exfiltrate all customer order records from Easy PayPal Events & Tickets plugin for WordPress through an exposed QR code scanning endpoint. The scan_qr.php file accepts sequential WordPress post IDs without authentication, enabling complete database harvesting of payment and customer information. Publicly available exploit code exists, but no evidence of active exploitation (not in CISA KEV). The plugin was officially closed and removed from WordPress.org on 2026-03-18, leaving existing installations vulnerable with no official patch path.

titra 0.99.52 leaks sensitive global configuration settings to any authenticated user via an unprotected Meteor DDP publication, exposing API keys and OAuth secrets without administrative checks. Authenticated attackers can subscribe to the globalsettings publication and retrieve plaintext credentials including google_secret, openai_apikey, and google_clientid. No public patch is available at time of publication.

OpenC3 COSMOS password change functionality accepts valid session tokens in lieu of current passwords, enabling attackers with hijacked tokens to lock out legitimate users and maintain persistent access to compromised accounts including administrator accounts. Publicly available exploit code demonstrates the attack chain. All versions prior to 6.10.5 and 7.0.0-rc1 through 7.0.0-rc2 are affected. The vendor has released patched versions 6.10.5 and 7.0.0-rc3 that enforce password-only verification during password change operations.

Information disclosure in Qualcomm Snapdragon firmware allows local authenticated attackers to read sensitive kernel memory via malformed IOCTL handler callbacks that bypass buffer size validation. The vulnerability affects multiple Snapdragon chipset versions and requires local access with limited privileges; exploitation results in confidentiality breach without direct system compromise. No active exploitation has been confirmed at the time of analysis.

CEL injection in Apache Polaris 1.4.0 allows authenticated users to escape credential access boundaries on Google Cloud Storage. Attackers can craft namespace or table identifiers containing single quotes and CEL fragments to break out of quoted strings in Credential Access Boundary conditions, escalating temporary table-scoped GCS credentials to effectively bucket-wide access. Confirmed in private testing: attackers obtained credentials intended for one table but successfully listed, read, created, and deleted objects across unrelated tables and external prefixes within the entire configured bucket. EPSS data not yet available for this recent CVE; CVSS 9.4 reflects critical confidentiality, integrity, and availability impact across both vulnerable and subsequent systems (scope changed).

Remote code execution via double-free memory corruption in Apache HTTP Server 2.4.66's HTTP/2 protocol implementation allows authenticated attackers to compromise server integrity and confidentiality with high impact. Vendor-released patch 2.4.67 addresses the issue. No public exploit or active exploitation confirmed at time of analysis, but SSVC framework rates technical impact as total, indicating complete system compromise potential.

HTTP response splitting in Apache HTTP Server 2.4.0 through 2.4.66 allows remote attackers to inject arbitrary HTTP headers and content when the server acts as a proxy to untrusted or compromised backend servers, enabling cache poisoning, session fixation, and cross-site scripting attacks. CVSS 6.5 (moderate) with network attack vector, no authentication required, and confirmed automatable exploitation per CISA SSVC framework. Vendor-released patch: version 2.4.67.

ILM Informatique OpenConcerto 1.7.5 stores sensitive passwords in plaintext, allowing authenticated local users to retrieve embedded credentials with low complexity. The vulnerability enables information disclosure of authentication data accessible via local file access, confirmed by CISA SSVC framework as having partial technical impact but no evidence of active exploitation.

Incorrect permission assignment in OpenConcerto 1.7.5 enables local authenticated users with UI interaction to modify critical binary files, potentially allowing privilege escalation or persistent code execution through binary replacement. CVSS score of 2.4 (Low) reflects the local-only attack vector and requirement for user interaction, though the vulnerability creates a persistent integrity risk for affected deployments.

Out-of-bounds read in mod_proxy_ajp of Apache HTTP Server through version 2.4.66 allows remote unauthenticated attackers to disclose sensitive information via a crafted AJP protocol request. The vulnerability has a CVSS score of 5.3 (moderate) with no active exploitation confirmed. Upgrade to version 2.4.67 to remediate.

Heap out-of-bounds read in Ollama's GGUF model loader (<0.17.1) leaks sensitive memory contents including API keys, environment variables, and concurrent user data when processing maliciously crafted model files. Attackers can trigger the vulnerability by uploading a GGUF file with tensor offsets exceeding file bounds via the unauthenticated /api/create endpoint, then exfiltrate leaked memory through /api/push to attacker-controlled registries. While default deployments bind to localhost only, the widely-adopted OLLAMA_HOST=0.0.0.0 configuration exposes instances to network exploitation. Vendor-released patch available in version 0.17.1 with GitHub commit 88d57d0483cca907e0b23a968c83627a20b21047 adding bounds validation to fs/ggml/gguf.go and server/quantization.go.

Mutt before 2.3.2 mishandles the IMAP GSS security level due to improper integer casting and insufficient bounds checking, allowing remote attackers to trigger memory corruption and information disclosure via a crafted IMAP server response during GSS-API authentication. The vulnerability requires high attack complexity (malicious IMAP server) but affects all versions prior to 2.3.2.

Magic Export & Import WordPress plugin before version 1.2.0 stores exported CSV files in a publicly accessible web directory, allowing unauthenticated remote attackers to enumerate and download sensitive user data without authentication. The vulnerability affects all versions prior to 1.2.0, has publicly available proof-of-concept code, and carries moderate real-world risk due to the low attack complexity and high automatable nature of exploitation, though the actual severity is constrained by the fact that CSV export must first occur and require that files remain accessible.

mutt before version 2.3.2 fails to validate null bytes during URL percent-decoding, allowing remote attackers to inject embedded null characters into decoded URLs, potentially causing information disclosure through truncation of validation checks or bypassing of security filters that rely on string length.

Out-of-bounds read in GoBGP BMP parser allows remote attackers to trigger information disclosure via malformed BMP messages to the BMPPeerUpNotification.ParseBody and BMPStatisticsReport.ParseBody functions. Affected versions are up to 4.3.0; patch available in version 4.4.0. CVSS 6.9 reflects availability impact. No public exploit code or active exploitation has been identified.

mutt before version 2.3.2 truncates the IMAP CRAM-MD5 authentication hash by one byte due to incorrect use of strfcpy instead of memcpy, potentially allowing attackers to bypass or weaken authentication on IMAP connections through off-by-one string handling errors.

Local privilege escalation in MediaTek geniezone component due to missing bounds check allows System-privileged actors to achieve total system compromise across multiple chipset models. The vulnerability requires prior System-level access and affects 17 MediaTek chipset variants (MT6899, MT8791T, MT8786, MT6789, MT8367, MT6768, MT8766, MT6993, MT6991, MT6877, MT8788E, MT8781, MT8768, MT6989, MT8910, MT8196, MT8793). No public exploit code identified at time of analysis; exploitation remains unconfirmed in active systems despite SSVC indicating total technical impact potential.

Mutt before 2.3.2 uses an unsafe string copy function (strfcpy) instead of memcpy when handling MD5 digest data in IMAP CRAM authentication, allowing attackers to potentially forge IMAP credentials by triggering buffer manipulation during the authentication handshake. The vulnerability requires manual connection attempt to a malicious IMAP server and affects network IMAP authentication flows, though the low CVSS score (3.7) reflects high attack complexity and integrity impact only.

Integer underflow in osrg GoBGP up to version 4.3.0 allows remote attackers to trigger a crash or information disclosure via crafted MRT (Multi-Threaded Routing Toolkit) packet data in the parseRibEntry function. The vulnerability arises from improper bounds checking when processing RIB (Routing Information Base) entries, enabling network-based exploitation without authentication. Vendor-released patch version 4.4.0 addresses this issue; no active exploitation has been confirmed at time of analysis.

Time-of-check time-of-use (TOCTOU) vulnerability in Prefect's validate_restricted_url function allows remote attackers with low privileges to bypass Server-Side Request Forgery (SSRF) protections via DNS rebinding attacks during webhook and notification delivery. Publicly available exploit code exists. The vulnerability affects Prefect versions up to 3.6.28.dev1 and is fixed in 3.6.28.dev2.

Authentication bypass in GeoVision LPC2011/LPC2211 license plate camera firmware version 1.10 allows remote unauthenticated attackers to brute-force predictable session cookies via the Web Interface and gain access to the device. The flaw stems from insufficient randomness in session token generation (CWE-341), enabling attackers to enumerate valid cookies through a crafted series of HTTP requests. No public exploit identified at time of analysis, and EPSS is low at 0.06%, but CISA SSVC rates technical impact as partial with no current exploitation observed.

Credential interception in GeoVision GV-IP Device Utility 9.0.5 allows network attackers to decrypt administrator passwords from broadcast UDP packets. The application broadcasts device commands with credentials encrypted using a modified Blowfish scheme, but includes the decryption key in the same packet - reducing security to algorithm obscurity. Attackers on the same LAN can capture these broadcasts when legitimate administrators interact with GeoVision IP cameras or devices, decrypt credentials using reverse-engineered algorithms, then gain full device control to reconfigure or factory-reset equipment. EPSS and KEV data not available; CVSS 9.3 reflects network-accessible credential disclosure requiring user interaction.

Improper authorization in Calibre-Web-Automated versions up to 4.0.6 allows authenticated remote attackers to generate or revoke Kobo authentication tokens for arbitrary users via an Insecure Direct Object Reference (IDOR) vulnerability in the kobo_auth.py component. An attacker with valid login credentials can request /kobo_auth/generate_auth_token/<victim_user_id> to obtain tokens authorizing Kobo sync as any other user, or revoke their tokens to deny service. Publicly available exploit code exists and the vulnerability is confirmed patched in version 4.0.7.

Assimp version 6.0.2 is vulnerable to denial of service through improper buffer handling in the FBXConverter::ConvertMeshMultiMaterial() function when processing crafted FBX model files. A remote attacker can trigger a crash by supplying a malicious FBX file via network or local file access, causing the application to become unavailable. Publicly available exploit code exists, though the vulnerability requires user interaction to process the malicious file.

Remote attackers can crash GoBGP v4.3.0 by sending a malformed BGP UPDATE message that triggers an out-of-bounds read in IPv6 extended community parsing. The flaw allows unauthenticated denial of service against default configurations with no authentication required (CVSS AV:N/AC:L/PR:N/UI:N). No public exploit identified at time of analysis, and EPSS score of 0.02% (4th percentile) indicates very low observed exploitation probability. Upstream fixes available via GitHub commits 362cce3e and 9ce8936.

Improper authorization in Calibre-Web up to version 0.6.26 allows authenticated remote attackers to manipulate the user_id parameter in the generate_auth_token function (cps/kobo_auth.py) to access or disclose information belonging to other users. The CVSS score of 2.1 reflects the requirement for prior authentication and limited impact scope, though publicly available exploit code exists.

Improper cryptographic signature verification in Dolibarr ERP CRM up to version 23.0.2 allows remote attackers to bypass signature validation in the Online Signature Module, potentially forging or manipulating signed transactions. The vulnerability affects the dol_verifyHash function and has been publicly disclosed with exploit code available, though exploitation requires high technical complexity and is not confirmed as actively exploited in production environments.

Frontend File Manager Plugin for WordPress through version 23.6 allows authenticated Subscriber-level users and higher to read arbitrary files belonging to other users via insecure direct object reference (IDOR) in the download endpoint. By manipulating the 'file_id' parameter, attackers can bypass authorization checks and access sensitive data stored by administrators and other privileged users. Publicly available exploit code exists for this vulnerability, though EPSS scoring (0.02%) suggests limited real-world exploitation relative to its high CVSS rating.

HTTP request smuggling in Starlet through version 0.31 allows remote unauthenticated attackers to bypass header validation by exploiting incorrect precedence of Content-Length over Transfer-Encoding headers. The vulnerability violates RFC 7230 section 3.3.3, which mandates that Transfer-Encoding must take precedence when both headers are present. An attacker positioned between a client and Starlet-based backend can craft malicious requests that are interpreted differently by a front-end reverse proxy and the Starlet server, enabling request smuggling attacks with integrity impact.

Improper restriction of excessive authentication attempts in CodeWise Tornet Scooter Mobile App 4.75 on iOS and Android allows remote attackers to bypass rate-limiting controls on the /TwoFactor endpoint, potentially enabling brute-force attacks against two-factor authentication mechanisms. Publicly available exploit code exists. The vendor did not respond to early disclosure notification.

Out-of-bounds read in MikroTik RouterOS 6.49.8 SCEP endpoint allows remote unauthenticated attackers to trigger memory disclosure and potential service disruption via malformed transactionID or messageType parameters. Public exploit code exists on GitHub. CVSS 7.3 reflects network-accessible attack surface with low complexity, though impact is rated limited across confidentiality, integrity, and availability. Vendor non-responsive to coordinated disclosure attempts.

Improper authorization in ChatGPTNextWeb NextChat up to version 2.16.1 allows remote unauthenticated attackers to bypass access controls in the addMcpServer function (app/mcp/actions.ts), potentially disclosing sensitive information. The vulnerability has a CVSS score of 5.5 (moderate severity) with public exploit code available, though the vendor has not yet responded to the disclosed issue.

Permissive cross-domain policy in ChatGPTNextWeb NextChat up to version 2.16.1 allows remote attackers with user interaction to manipulate the Next.js API endpoint and enable untrusted domains to access application resources, leading to information disclosure. Public exploit code exists for this vulnerability, though the vendor has not yet responded to early disclosure notification.

External file inclusion in Totolink N300RH firmware 6.1c.1353_B20190305 allows remote unauthenticated attackers to manipulate the FileName parameter in the setUploadSetting function via /cgi-bin/cstecgi.cgi, enabling arbitrary file writes and denial of service. Publicly available exploit code exists, and the vulnerability carries a CVSS score of 6.5 reflecting network-accessible attack vector with low complexity.

Unauthenticated remote attackers can extract sensitive user information including email addresses, usernames, and user IDs through the '/dokan/v1/stores/{id}/reviews' REST API endpoint in Dokan plugin versions up to 4.3.1. The vulnerability affects only installations with the Pro version activated and store reviews enabled, allowing information disclosure of all customers who left vendor reviews. No public exploit code has been identified, but the attack requires no authentication or user interaction and can be automated at scale.

Insufficient verification of firmware authenticity in TRENDnet TEW-821DAP up to version 1.12B01 allows remote attackers to manipulate firmware updates through the cameo_dev.sh update handler, potentially leading to information disclosure. The vulnerability requires high attack complexity and difficult exploitation conditions. The affected hardware (v1.xR) reached end-of-life eight years ago and is no longer supported by the vendor, significantly limiting real-world exposure.

TRENDnet TEW-821DAP firmware version 1.12B01 transmits sensitive information in cleartext via the /www/cgi/ssi endpoint during firmware updates, allowing remote unauthenticated attackers to intercept credentials or configuration data. The vulnerability affects end-of-life hardware (version v1.xR) discontinued 8 years ago and no longer supported by the vendor. Public exploit code is available, though the attack requires high complexity conditions to execute successfully.

Insufficient verification of data authenticity in the firmware update handler of TRENDnet TEW-821DAP 1.12B01 allows remote attackers to manipulate the dest argument during firmware updates, leading to integrity violations. The vulnerability requires high attack complexity and affects only end-of-life hardware (version 1.xR) discontinued 8 years ago. CVSS score is 3.7 (low) with integrity impact but no confidentiality or availability impact; no public exploit confirmed.

My Social Feeds - Social Feeds Embedder WordPress plugin up to version 1.0.4 exposes sensitive TikTok OAuth credentials via an unauthenticated AJAX endpoint due to missing authorization and nonce checks. Authenticated attackers with Subscriber-level access can retrieve stored access_token and refresh_token values belonging to administrator-connected TikTok accounts, enabling them to impersonate the site owner in TikTok API interactions. No public exploit code or active exploitation has been reported at the time of analysis, though the vulnerability is trivial to exploit once network access is established.

Unauthenticated attackers can access and modify plugin settings in the Widgets for Social Photo Feed WordPress plugin through missing capability checks on two REST API endpoints, allowing unauthorized data access and configuration changes in all versions up to and including 1.8. The vulnerability requires only network access with no user interaction, making it trivially exploitable by remote attackers without authentication credentials.

Improper authorization in JeecgBoot up to version 3.9.1 allows authenticated remote attackers to manipulate the ruleClass parameter in the /sys/fillRule/edit endpoint, leading to unauthorized access and information disclosure. The vulnerability affects the FillRuleUtil component and has publicly available exploit code; the vendor confirmed the issue and committed to patching in an upcoming release.

Authenticated users with low privileges can read cleartext Kubernetes Secret data through Argo CD's ServerSideDiff feature in versions 3.2.0-3.2.10 and 3.3.0-3.3.8. This scope-changing vulnerability (CVSS:3.1 S:C) allows attackers to access sensitive credential data managed by Kubernetes, including database passwords, API tokens, and certificates, by exploiting the server-side diff functionality. With a 7.7 CVSS score and low attack complexity (AC:L), this represents a significant confidentiality breach requiring only network access and basic authentication-no public exploit identified at time of analysis, but the technical barrier to exploitation is minimal.

HTTP request smuggling in mtrudel bandit before version 1.11.0 allows unauthenticated attackers to bypass edge security controls when the application sits behind a proxy that interprets duplicate Content-Length headers differently. The vulnerability stems from Bandit accepting only the first Content-Length header while proxies may use the last value, causing request framing desynchronization that enables smuggling past WAF rules, path-based ACLs, rate limiting, and audit logging. CVSS 6.3 (AV:N/AC:L/AT:P) indicates network-accessible exploitation with some attack timing complexity; no public exploit code or active KEV listing identified at analysis time, but RFC 9112 non-compliance creates a known attack pattern.

DTrace kernel instrumentation tool on Linux is vulnerable to a denial-of-service and potential privilege escalation attack when processing malicious ELF binaries with out-of-range sh_link fields. An unprivileged attacker can craft an ELF binary that, when instrumented by a root-level dtrace process, triggers an out-of-bounds heap read in the ELF parser. This can crash the dtrace daemon (DoS) or, depending on heap layout, lead to reading and dereferencing garbage pointers controlled by the attacker, potentially enabling code execution in a privileged context. The vulnerability requires local access and a privileged dtrace instance to be actively instrumentation the malicious process, but carries significant risk given dtrace's typical deployment in system administration and security monitoring contexts.

Incomplete validation of AI rich response messages for Instagram Reels in WhatsApp for iOS v2.25.8.0 through v2.26.15.72 and Android v2.25.8.0 through v2.26.7.10 allows authenticated attackers to trigger processing of media content from arbitrary URLs on a victim's device, potentially invoking OS-controlled custom URL scheme handlers to disclose limited information. No active exploitation has been observed in the wild, and CISA SSVC indicates no known exploitation path despite the network attack vector.

WhatsApp for Windows prior to v2.3000.1032164386.258709 permits attachment spoofing via maliciously formatted documents with embedded NUL bytes in filenames, causing the application to display files as benign types while executing them as executables upon opening. The vulnerability requires user interaction to open a crafted attachment delivered over the network, enabling an attacker to achieve code execution with the privileges of the WhatsApp process. No public exploit code or active exploitation has been confirmed at time of analysis.

Unauthenticated traffic relay vulnerability in Prosody mod_proxy65 module allows network attackers to bypass access control during SOCKS5 activation, resulting in integrity compromise and service disruption without requiring authentication. Affected versions are Prosody before 0.12.6 and 1.0.0 through 13.0.0 before 13.0.5 when mod_proxy65 is enabled. CVSS 6.5 reflects medium severity with network-accessible attack surface, low complexity, and non-privileged unauthenticated access, though confirmed availability and integrity impact limits widespread severity.

Use-after-free in Linux kernel ksmbd (SMB server) during durable file handle scavenging allows memory corruption and potential remote code execution. When a durable SMB2 file handle survives session disconnect, the cleanup path dereferences a freed connection object via NULL fp->conn pointer during lock cleanup, causing a slab use-after-free. Exploitation probability is extremely low (EPSS 0.02%, 5th percentile) with no active exploitation confirmed. Vendor patches available across multiple stable kernel branches (6.12.84, 6.18.25, 7.0.2, 7.1-rc1) address the asymmetric cleanup by properly managing byte-range lock lifetimes during durable handle reconnection.

Authentication bypass in Linux kernel's ksmbd SMB server allows any authenticated SMB user to hijack orphaned durable file handles by predicting persistent IDs, enabling unauthorized file access with the original owner's privileges. The flaw violates MS-SMB2 requirements for SecurityContext validation during durable handle reconnection. Vendor patches are available across multiple stable kernel branches (6.18.25, 7.0.2, 7.1-rc1). EPSS exploitation probability is low at 0.02% (4th percentile), and no active exploitation or public POC has been identified. CVSS 8.8 reflects high impact but requires low-privilege authentication.

Use-after-free in Linux kernel F2FS filesystem allows local authenticated attackers to trigger kernel panic or potentially achieve code execution. The vulnerability (CWE-416) occurs during concurrent write callback and unmount operations when f2fs_write_end_io() decrements page count before checking node inode validity, leading to NULL pointer dereference. Discovered via xfstests generic/107 and syzbot fuzzing. EPSS exploitation probability is low (0.02%, 4th percentile), no active exploitation confirmed. Vendor patches available across stable kernel branches 6.18.25, 7.0.2, and 7.1-rc1.

Memory leak in f2fs_rename() function allows local authenticated attackers to cause denial of service through repeated file rename operations. The vulnerability exists in the f2fs filesystem implementation when handling SELinux label initialization during whiteout file creation, due to a missing f2fs_free_filename() call introduced in commit 40b2d55e0452. Vendor patches are available for Linux 6.6.136, 6.12.84, 6.18.25, 7.0.2, and 7.1-rc1.

Resource exhaustion in Linux kernel ksmbd server allows remote unauthenticated attackers to permanently deny SMB service by consuming connection slots through forced allocation failures. The vulnerability leaks active_num_conn counter values when alloc_transport() fails during TCP connection setup on port 445, permanently consuming slots from the max_connections pool until module reload. Attackers can accelerate exhaustion by holding open connections with large RFC1002 lengths (up to 16MB) to trigger memory pressure. EPSS score of 0.11% suggests low observed exploitation probability, and no public exploit or CISA KEV listing exists. Vendor patches available for kernel versions 6.6.136, 6.12.84, 7.0.2, and 7.1-rc1.

In the Linux kernel, the following vulnerability has been resolved: smb: client: fix dir separator in SMB1 UNIX mounts When calling cifs_mount_get_tcon() with SMB1 UNIX mounts, @cifs_sb->mnt_cifs_flags needs to be read or updated only after calling reset_cifs_unix_caps(), otherwise it might end up with missing CIFS_MOUNT_POSIXACL and CIFS_MOUNT_POSIX_PATHS bits. This fixes the wrong dir separator used in paths caused by the missing CIFS_MOUNT_POSIX_PATHS bit in cifs_sb_info::mnt_cifs_flags.

Remote unauthenticated attackers can trigger out-of-bounds memory access in the Linux kernel SMB client's DACL parsing code by sending a malicious SMB response with a truncated DACL structure. The vulnerability exists in build_sec_desc() and id_mode_to_cifs_acl() functions which insufficiently validate server-supplied ACL data before rewriting it during chmod/chown operations, allowing ACE traversal beyond validated memory bounds. CVSS 8.8 indicates high severity with network vector requiring user interaction. EPSS score of 0.02% (5th percentile) suggests low observed exploitation probability in the wild, and no active exploitation is confirmed (not in CISA KEV). Vendor patch available targeting Linux kernel 7.0.2 and 7.1-rc1.

A malicious SMB server can trigger out-of-bounds heap memory disclosure in Linux kernel SMB client (CIFS) through crafted QUERY_INFO responses. Vulnerable Linux kernel versions 5.1 through 6.12.84 do not validate server-reported OutputBufferLength against actual response size before copying data to userspace, allowing a rogue SMB server to expose adjacent kernel heap contents. Patches available across stable kernel branches (6.6.136, 6.12.84, 6.18.25, 7.0.2, 7.1-rc1). EPSS score of 0.02% indicates low exploitation probability; no active exploitation confirmed. Attack requires user interaction to mount malicious SMB share.

Use-after-free in Linux kernel f2fs compressed writeback allows local authenticated users to trigger memory corruption, potentially executing arbitrary code or causing system crashes. Affects f2fs-compressed filesystems in Linux kernel 5.6 through 7.1-rc2, with patches available in 6.6.136, 6.12.84, 7.0.2, and 7.1-rc1. EPSS score of 0.02% (5th percentile) indicates low observed exploitation probability despite CVSS 7.8 rating. This mirrors CVE-2026-23234's race condition pattern but in the compression code path that was missed by the earlier fix. No active exploitation confirmed (not in CISA KEV) and no public POC identified at time of analysis.

Use-after-free vulnerability in the ALSA caiaq USB audio driver allows local authenticated attackers to cause denial of service by triggering asynchronous card free callbacks after USB device disconnection. The vulnerability stems from missing reference counting on the parent USB device pointer, combined with an inappropriate usb_reset_device() call in the card teardown path. EPSS exploitation probability is minimal (0.02%), and no public exploit code or active exploitation has been identified.

Use-after-free in Linux kernel virt_wifi driver allows local authenticated users to trigger memory corruption during ethtool operations on virtual WiFi devices being unregistered. The vulnerability stems from improper device parent reference handling via SET_NETDEV_DEV, where ethnl_ops_begin() calls pm_runtime_get_sync() on already-freed memory when a virt_wifi device unregisters concurrently with ethtool operations. Patches are available across multiple stable kernel branches (5.15.203, 6.1.168, 6.6.134, 6.12.81, 6.18.22, 6.19.12, 7.0). EPSS exploitation probability is low (0.02%, 7th percentile), and no public exploit identified at time of analysis, though CVSS 7.8 reflects potential for complete system compromise if successfully triggered.

A denial-of-service vulnerability in the Linux kernel's IPv6 checksum GSO fallback logic allows remote unauthenticated attackers to trigger system instability via specially crafted tunneled IPv6 packets with extension headers. The flaw affects network packet processing when NETIF_F_IPV6_CSUM offload is enabled, causing incorrect handling of tunneled traffic that requires software checksumming. EPSS score is low (0.02%, 7th percentile), indicating minimal observed exploitation activity. Vendor-released patches are available across multiple kernel version branches (6.1.168, 6.6.134, 6.12.81, 6.18.22, 6.19.12, 7.0), confirmed by upstream Linux kernel commits. No public exploit identified at time of analysis, though CVSS vector indicates straightforward network-based exploitation (AV:N/AC:L/PR:N/UI:N).

Use-after-free in Linux kernel's MANA network driver allows local authenticated attackers to corrupt memory and potentially execute code with kernel privileges. The flaw occurs when auxiliary_device_add() fails in add_adev(), triggering cleanup that frees memory still referenced by subsequent error-handling code. Patches available across stable kernel branches (6.6.134, 6.12.81, 6.18.22, 6.19.12, 7.0). EPSS score of 0.02% (5th percentile) indicates low probability of widespread exploitation. No CISA KEV listing or public exploit identified at time of analysis.

Uninitialized memory in the Linux kernel SCSI target subsystem (target_core_file) causes write operations to fail unpredictably when bogus ki_write_stream values trigger block device validation checks. Affected versions span Linux kernel 6.16 through development branches, with stable patches released for 6.18.22, 6.19.12, and 7.0. While CVSS scores this as 7.5 High with network vector (AV:N), the description indicates a local kernel subsystem issue affecting SCSI target configurations, suggesting a vector/impact mismatch. EPSS probability is very low (0.02%, 4th percentile) with no evidence of active exploitation or public POC, indicating minimal real-world targeting despite the high CVSS score.

tcm_loop target reset handler fails to drain in-flight SCSI commands, violating SCSI error handling contract and causing LUN reference leaks that deadlock configfs LUN unlink operations. Local users with appropriate privileges can trigger denial of service by initiating reset sequences while SCSI commands are in flight, leaving the kernel in an unkillable D-state waiting for LUN reference counts to clear. This is a local denial of service affecting the SCSI target core's tcm_loop loopback driver across multiple kernel versions.

Logic error in Linux kernel mac80211 TDLS handling allows local authenticated users to modify wireless channel context and HT protection settings by invoking NL80211_TDLS_ENABLE_LINK on non-TDLS stations. Missing validation causes the kernel to apply TDLS-specific operations to regular Wi-Fi stations, potentially disrupting wireless connectivity and creating integrity/availability impacts. Vendor patches available for kernel 6.12.81, 6.18.22, 6.19.12, and 7.0. EPSS score of 0.02% (5th percentile) indicates very low probability of exploitation despite CVSS 7.1 rating. No evidence of active exploitation or public POC identified at time of analysis.

An out-of-bounds read vulnerability in the Linux kernel's Wacom HID driver (wacom_intuos_bt_irq function) allows adjacent network attackers to cause information disclosure or denial of service through maliciously crafted short Bluetooth HID reports. The vulnerability affects the Bluetooth interface of Wacom Intuos tablets, where report types 0x03 and 0x04 are processed without validating minimum lengths (22 and 32 bytes respectively), enabling memory reads beyond buffer boundaries. Patches are available across multiple stable kernel versions (5.10.253, 5.15.203, 6.1.168, 6.6.134, 6.12.81, 6.18.22, 6.19.12, 7.0) with no active exploitation confirmed (EPSS 0.02%, not in CISA KEV).

Use-after-free in Linux kernel HID subsystem allows local attackers with low privileges to achieve arbitrary code execution, privilege escalation, or denial of service when force feedback initialization fails on Logitech G920 racing wheels. The vulnerability occurs when userspace continues accessing freed memory structures (sysfs and /dev/input) after initialization errors. Vendor patches available across multiple stable kernel branches (6.12.81, 6.18.22, 6.19.12, 7.0). EPSS score of 0.02% (5th percentile) indicates very low probability of mass exploitation, consistent with hardware-specific local attack surface requiring physical device presence.

Adjacent network attackers can achieve high-severity code execution, information disclosure, or denial of service in the Linux kernel HID (Human Interface Device) subsystem by exploiting a bounds-checking flaw in hid_report_raw_event(). A bogus memset() operation intended to zero unused buffer space instead creates out-of-bounds read/write conditions when processing malformed HID input reports from adjacent devices (USB, Bluetooth). Vendor patches available for stable branches 6.18.22, 6.19.12, and 7.0. EPSS score of 0.02% suggests minimal observed exploitation, but the unauthenticated adjacent-network attack vector with low complexity makes this exploitable in environments with untrusted HID peripherals.

Kernel denial of service via crafted btrfs metadata allowing local attackers to trigger an unguarded BUG_ON() condition during relocation recovery at mount time. The vulnerability arises when a root item on disk contains a non-zero drop_progress with zero drop_level, an invalid state that should not exist but lacks validation on read. CVSS 5.5 reflects local attack vector and availability impact; EPSS 0.02% indicates minimal real-world exploitation likelihood.

Memory corruption in the Linux kernel CAAM (Cryptographic Acceleration and Assurance Module) crypto driver allows local authenticated users to corrupt kernel memory and potentially escalate privileges. The vulnerability occurs when HMAC keys longer than the hash block size are processed - the driver allocates a DMA-aligned buffer size but fails to use it, causing the hashed key to overwrite adjacent memory. Vendor patches are available for stable kernel versions 6.6.134, 6.12.81, 6.18.22, 6.19.12, and 7.0. EPSS score of 0.02% (5th percentile) indicates very low observed exploitation probability, and no public exploit code or CISA KEV listing exists at time of analysis.

Race condition in Linux kernel MPLS subsystem allows local authenticated users to trigger out-of-bounds memory access via concurrent label table resizing. The vulnerability affects RCU-protected codepaths (mpls_forward, mpls_dump_routes) that can obtain inconsistent snapshots of platform_labels array metadata during resize operations, potentially leading to information disclosure or denial of service. Vendor patch available addressing the issue through seqcount-based synchronization. EPSS score of 0.02% (5th percentile) indicates very low observed exploitation probability, and no active exploitation or public POC identified.

In the Linux kernel, the following vulnerability has been resolved: net: qrtr: replace qrtr_tx_flow radix_tree with xarray to fix memory leak __radix_tree_create() allocates and links intermediate nodes into the tree one by one. If a subsequent allocation fails, the already-linked nodes remain in the tree with no corresponding leaf entry. These orphaned internal nodes are never reclaimed because radix_tree_for_each_slot() only visits slots containing leaf values. The radix_tree API is deprecated in favor of xarray. As suggested by Matthew Wilcox, migrate qrtr_tx_flow from radix_tree to xarray instead of fixing the radix_tree itself [1]. xarray properly handles cleanup of internal nodes - xa_destroy() frees all internal xarray nodes when the qrtr_node is released, preventing the leak. [1] https://lore.kernel.org/all/20260225071623.41275-1-jiayuan.chen@linux.dev/T/