Stored cross-site scripting (XSS) in the Crocoblock JetSmartFilters WordPress plugin through version 3.8.6 lets an authenticated low-privileged user (CVSS PR:L, e.g., contributor or author) inject a persistent payload into filter-handled input that is improperly neutralized before web page generation, and the script then executes in the browser of a higher-privileged victim such as an administrator who views the affected page (UI:R). The issue is scored CVSS 6.5 (AV:N/AC:L/PR:L/UI:R/S:C/C:L/I:L/A:N per our assessment), is a moderate-severity CWE-79 defect, and there is no public exploit code or confirmed active exploitation identified at time of analysis; it is not exploitable against a default site without a valid account on the target WordPress installation.
Missing authorization in the mail rules feature of JetBrains YouTrack allows an authenticated user holding project-administrator privileges to execute issue commands against projects they do not administer. The flaw (CWE-862) is reachable remotely over the network with low attack complexity, but it is privilege-gated: the attacker must already hold project-admin rights and the instance must process mail rules, so exposure is concentrated in multi-team or multi-tenant deployments where project administration is delegated. Impact is a serious integrity and confidentiality concern for affected projects (CVSS 6.5), but there is no public exploit identified at time of analysis and no indication of active exploitation. Fixed builds are available in the 2026.2 and 2026.3 release lines.
Out-of-bounds read in HPE NonStop SQL/MX EXE allows local attackers with low-level privileges to cause information disclosure, process termination, CPU halt, or data corruption by supplying a corrupted file. This medium-severity vulnerability (CVSS 6.5) requires a specific file state to trigger and has no public exploit identified at time of analysis. It affects HPE NonStop SQL/MX EXE deployments.
Stored cross-site scripting in the ThemeGrill User Registration WordPress plugin (versions through 5.2.8) allows an authenticated user with access to a plugin input field that persists data to store a malicious script which later executes in the browser of any other user viewing the affected page, including administrators. Rated CVSS 6.5 (CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:L/I:L/A:L), exploitation requires an authenticated low-privileged account plus subsequent victim interaction, and no public exploit code has been identified at time of analysis. Impact is bounded to the victim's browser session rather than direct server compromise, but the scope change (S:C) means the injected script can act with the victim's privileges against the surrounding application.
Code injection in the WPMU DEV Broken Link Checker WordPress plugin, affecting all versions up to and including 2.4.14.1, allows an unauthenticated remote attacker to feed crafted input to the plugin's link-checking/parsing routine and potentially expose sensitive data. Per the assessed CVSS vector (AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:N/A:N), exploitation is network-reachable and needs no privileges, but a victim must be induced to trigger the vulnerable action, and the confirmed impact is confined to confidentiality with no integrity or availability effect. No public exploit identified at time of analysis.
Rate limiting in Eclipse Foundation's api-gateway component of the Eclipse Connected Services Platform (versions up to and including 1.7.3) can be completely bypassed by any remote client that varies the X-Forwarded-For header, because the shipped default ClientIpKeyResolver builds each rate-limit bucket key from the unvalidated leftmost element of that client-supplied header, letting one attacker mint an unlimited number of independent buckets and exceed the configured request-rate limits on every route using the default configuration. The vulnerability is a medium-severity (CVSS 6.5) reliance-on-untrusted-source flaw (CWE-348) whose impact is availability-oriented, enabling resource exhaustion and abuse of downstream services; although the vendor vector scores it PR:L, the independent assessment rates it PR:N because spoofing the header requires no authentication at all. There is no CISA KEV entry and no public exploit code identified at time of analysis, but an upstream fix is already public in the project's pull request #330, which shortens the window for attackers to locate and target unpatched, default-configured deployments. Exploitation is conditional: it applies only where the default key resolver is in use, where the attacker can set arbitrary X-Forwarded-For values, and where no trusted-proxy allowlist validates that header.
Stored cross-site scripting in the VK Blocks WordPress plugin (all versions up to and including 1.126.1) allows an authenticated user with contributor-level privileges or above to inject persistent JavaScript through the Group Block's 'linkToPost' feature, which then executes in the browser of anyone who opens the affected page - including higher-privileged editors and administrators who review the content. The defect is insufficient input sanitization combined with missing output escaping (CWE-79), rated CVSS 3.1 6.4 (AV:N/AC:L/PR:L/UI:N/S:C/C:L/I:L/A:N), with scope change but only low confidentiality and integrity impact and no server-side compromise. No public exploit code was identified at time of analysis and the issue is not listed in CISA KEV, so real-world priority is tempered by the requirement to first obtain a contributor account.
Stored cross-site scripting in the WordPress plugin "Import and export users and customers" (all versions through 2.4.5) lets authenticated attackers holding only custom-level access inject persistent JavaScript through Custom Column User Meta Profile Fields, which then executes whenever a page rendering that meta is viewed - typically by a higher-privileged administrator reviewing user records. The Wordfence-assigned CVSS of 6.4 (AV:N/AC:L/PR:L/UI:N/S:C/C:L/I:L/A:N) reflects a low-privilege, network-reachable flaw with limited confidentiality and integrity impact but a scope change, because script executes in the context of the reviewing user's session rather than the attacker's own. Exploitation is constrained by the need for prior authentication, an account permitted to write custom user meta, and a victim subsequently loading the injected page, so sites that run this plugin and issue accounts to untrusted users (membership sites, multi-author blogs, customer-facing stores) carry the real risk; no public exploit code was identified at time of analysis and the issue is not listed in CISA KEV.
Stored cross-site scripting in the Kirki - Freeform Page Builder, Website Builder & Customizer plugin for WordPress (all versions up to and including 6.2.0) lets an authenticated user who has been granted the administrator-controlled kirki_content_access capability save arbitrary JavaScript through the Page Builder Text Content field, which then executes in the browser of any user who later views the injected page. Exploitation requires that capability grant, which is a non-default, elevated content permission rather than a standard subscriber or author role, and the stored payload only fires on a subsequent victim page view; upgrading to 6.2.1 removes the flaw. No public exploit code, proof-of-concept, or CISA KEV entry has been identified at the time of analysis, and this rates as a moderate, not high-urgency, issue.
Stored cross-site scripting in the BoldGrid 'Post and Page Builder - Visual Drag and Drop Editor' WordPress plugin (all versions up to and including 1.27.13) lets an authenticated attacker with Contributor-level access or higher plant arbitrary JavaScript inside the Table Designer's 'data-label' attribute, which is stored without adequate sanitization or output escaping and then executes in the browser of any user who later views the injected page. Because the payload is persistent and rendered in the site's origin, successful exploitation can lead to session theft, content tampering, or privilege escalation for the viewing user; CVSS 3.1 score is 6.4 (Medium) with vector CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:L/I:L/A:N. No public exploit identified at time of analysis and no CISA KEV listing; vendor fix is available in version 1.27.14. Sites that restrict content authoring to trusted users are substantially less exposed.
Out-of-bounds memory read and write operations in HPE NonStop EXPAND can be triggered by malformed traffic sent to the EXPAND networking subsystem, potentially leading to information disclosure or memory corruption. The vulnerability is rated CVSS 6.4 (AV:A/AC:H/PR:N/UI:N) - an attacker needs an adjacent-network position and no credentials, but exploitation depends on specific, attacker-uncontrolled memory or timing conditions, which keeps this at moderate rather than critical severity. No public exploit code or active exploitation has been identified at time of analysis; remediation guidance should be taken from the HPE security bulletin.
Mutation cross-site scripting in the ammonia Rust HTML sanitizer (versions prior to 3.3.2, 4.0.2, and 4.1.3) allows a remote, unauthenticated attacker to execute JavaScript in a victim's browser when the victim renders sanitized output. The flaw is config-gated: it only triggers for applications that explicitly enable the math tag, the annotation-xml tag, and at least one raw-text HTML element (such as xmp) while disallowing the encoding attribute, which is a non-default allowlist - stock ammonia configurations are not affected. ammonia's own advisory confirms the issue and vendor fixes ship in 3.3.2, 4.0.2, and 4.1.3; there is no confirmed active exploitation (not in CISA KEV) and no public exploit code identified at time of analysis in the supplied intelligence, though a public technical writeup of the mutation technique is referenced.
Memory corruption and null pointer dereference in the HPE NonStop SQL/MX Run-Time Statistics component let an attacker who already holds low-level authenticated access to a NonStop system (PR:L) crash the affected process, producing a denial of service, with a secondary possibility of limited information disclosure (C:L, I:L). The high attack-complexity rating (AC:H) reflects that a single deterministic request is not enough - the attacker must craft malformed input that satisfies specific state or timing conditions before the flaw triggers. This is a moderate-severity (CVSS 6.4) input-validation defect (CWE-20) reported by HPE itself; no public exploit code and no confirmed active exploitation were identified at time of analysis, and exposure is limited to NonStop environments that actually run the SQL/MX Run-Time Statistics feature.
Reflected cross-site scripting in Netgate pfSense CE 2.7.1 through 2.9.0 allows unauthenticated remote attackers to inject script into the WebGUI via the unencoded 'if' parameter in firewall_rules.php, but only when the non-default 'Require Firewall Interface' option is enabled and a privileged operator is lured into opening a crafted GET link. Publicly available exploit code exists, yet the vulnerability is a low-priority issue because exploitation requires a non-default configuration, user interaction, and a victim who already holds firewall-rules privileges, with impact confined to that victim's session.
Authenticated cross-site scripting in QNAP's QuFTP Service allows a remote attacker who already holds a valid QuFTP user account to inject malicious script that executes in a victim's browser within the NAS web interface, potentially compromising the wider session context rather than the FTP service alone. The flaw affects QuFTP Service installations on QNAP NAS devices in branches prior to 1.4.6, 1.5.5, and 1.6.5, and successful exploitation additionally requires a victim to view the injected content. No public exploit code and no confirmed active exploitation were identified at time of analysis, and the mandatory account prerequisite plus the user-interaction requirement are the principal limiters on real-world risk.
Server-side request forgery in web3.py's CCIP Read (EIP-3668) URL handling (CVE-2026-111171) allows a malicious or attacker-controlled smart contract to make victim applications issue outbound HTTP requests to loopback, RFC1918, or link-local destinations, including cloud metadata services such as 169.254.169.254. The flaw affects web3.py 7.15.0 through 7.16.0 and 8.0.0b2 through 8.0.0, and is reachable unauthenticated (CVSS:4.0 PR:N, AV:N) by any backend, indexer, bot, or API that performs .call()/eth_call against untrusted contracts, since CCIP Read is enabled by default (global_ccip_read_enabled=True). Exploitation requires the attacker to deliberately craft an IPv4-mapped IPv6 host (e.g. ::ffff:127.0.0.1) or AAAA record that bypasses _check_ip_blocked(), so the attack is automatable only with specific crafting (SSVC Automatable: no); publicly available exploit code exists, impact is partial (low confidentiality and integrity), and the issue is not currently listed in CISA KEV.
OxiCloud 0.9.1 through 0.9.3 fails to revoke previously issued anonymous share grants when a share owner turns on password protection, so anyone still holding an oxi_shares share-ring cookie can continue reading shared files and folder listings through GET /api/files/{id} (CVSS 6.3, Medium; CWE-863). Exploitation is unauthenticated (PR:N) but gated by an attack requirement (AT:P): the attacker must have already accessed that specific share while it was public and must still possess the resulting cookie, and the resulting impact is limited to disclosure of shared content with no integrity or availability effect. No public exploit identified at time of analysis.
Authenticated users with low-privileged WordPress accounts can exploit an insecure direct object reference (IDOR) in the Simply Schedule Appointments plugin through version 1.6.12.35 to read and alter appointment and booking records belonging to other users, because the plugin's endpoints accept a user-controlled object key without verifying ownership. The flaw requires valid credentials (CVSS vector PR:L, so it is not pre-auth) and yields only partial confidentiality and integrity impact rather than full site takeover; no public exploit code has been identified at time of analysis and the issue is not listed in CISA KEV, making this a moderate, authenticated-user authorization weakness rather than an emergency. Sites that grant subscriber-level or other low-privilege accounts are the primary exposure surface.
Server-side request forgery via DNS rebinding in the AliasVault self-hosted server's favicon extractor (versions prior to 0.30.3) allows an authenticated, low-privileged user to make the server issue HTTP/HTTPS requests to internal hosts and read back limited information about them. The flaw is a time-of-check/time-of-use race (CWE-367): the API validates a user-supplied hostname with one DNS lookup, but the HTTP client performs a second, independent lookup when connecting, so an attacker-controlled authoritative DNS server can answer with a public address during validation and a private address during connection. Exploitation is conditional and moderate in severity rather than urgent - it requires a publicly reachable self-hosted instance with public registration enabled, a valid AliasVault account, an attacker-controlled authoritative DNS server, and a resolver that permits rebinding; default checks limit outbound requests to ports 80 and 443, and the browser extensions, web application, and mobile apps are not affected. No public exploit code has been identified at time of analysis and the issue is not confirmed as actively exploited (no CISA KEV listing); the vendor released a fix in 0.30.3.
Rancher Fleet's default SSH host-key validation trusts unverified server certificates when no known_hosts entry is present, enabling an on-path attacker to conduct a man-in-the-middle attack against SSH connections initiated by Fleet's git fetcher, cloner, or fleet apply command. Exploitation requires the attacker to occupy an on-path network position and the target server to lack a trusted known_hosts entry, yielding only low-level confidentiality, integrity, and availability impact. This is a moderate-priority issue, not a drop-everything emergency; no public exploit has been identified at time of analysis, and vendor-released patches are available in versions 0.10.12, 0.11.7, and 0.12.2.
Local low-privileged users on HPE NonStop systems can potentially read privileged or unmapped kernel memory through improper handling of overlapping protected memory ranges, resulting in disclosure of sensitive data. Exploitation requires an attacker to already hold authenticated local execution on the NonStop host and to satisfy high-complexity conditions such as a specific kernel state, timing, or memory layout, so the issue is not reliably reproducible on demand and has no network-facing entry point. This is a moderate, not top-priority, issue: no public exploit code has been identified at time of analysis, and the practical risk is confined to insiders or post-compromise escalation on this specialized platform.
Stored cross-site scripting in QNAP's Network & Virtual Switch component, shipped with QTS, QuTS hero, and QuTS cloud, lets an attacker who already holds a QNAP administrator account inject malicious script into the management interface, which then executes in the browser of another administrator who loads the affected page. Exploitation therefore requires two preconditions that are both confirmed by the CVSS vector (PR:H and UI:P): valid administrative credentials and victim interaction, which confines realistic impact to multi-administrator deployments where attackers can reach an already-authenticated peer. No public exploit code and no confirmed active exploitation were identified at time of analysis, and the CVSS 4.0 base score is only 2.2, so this is a low-priority, post-compromise hardening issue rather than an externally exploitable entry point.
A stack-based buffer overflow in the gensub_ParsePsb() subtitle parser of AOC LC32D1320 firmware v2.05 lets an attacker crash the display by supplying a crafted .PSB subtitle file. The vector is local (AV:L) with no privileges or user interaction required in the CVSS scoring (PR:N/UI:N), and impact is limited to availability (A:H) - a denial of service, not code execution or data exposure - giving a medium severity of 6.2. No CISA KEV entry exists and EPSS is minimal (0.17%, 6th percentile), so real-world exploitation is not confirmed; a public technical writeup and an associated libaoc code repository are linked, but working exploit code is not verified. Priority guidance places remediation in the this_month window rather than immediate action.
Authentication bypass in Scoold (Erudika) prior to 1.69.1 allows an attacker who already possesses one of the server signing secrets - app_secret_key or para_secret_key - to forge a JWT that is accepted for the /api/config and /api/stats endpoints and mapped to the hardcoded administrative API_USER, thereby reading sensitive configuration and server logs and reaching privileged configuration functionality. An ordinary user login-session token is explicitly insufficient for this attack, and the vulnerable check runs even when scoold.api_enabled is false, so turning off the API is not a mitigation. No public exploit code and no confirmed active exploitation were identified at time of analysis; because the attack hinges entirely on obtaining a server-side signing secret, the input CVSS of 6.1 (assessed vector AV:N/AC:L/PR:H/UI:N/S:U/C:H/I:H/A:N) reflects a medium-severity, heavily preconditioned issue rather than an open, remotely reachable bypass.
Cross-course submission tampering in ClassroomIO through 1.0.0 lets an authenticated course tutor or admin overwrite answers, grades, feedback and grading status - or delete submissions outright - in courses they are not assigned to teach. The Hono API PUT and DELETE submission handlers accept a caller-supplied course id alongside a submission UUID and never verify that the submission actually belongs to that course (CWE-639), so any staff account holding tutor/admin privileges in at least one course can reach other courses' submissions if it knows the target UUID. Exploitation therefore requires authenticated privileged access plus a random, non-enumerable submission UUID (AT:P), and there is no anonymous or unauthenticated path; no public exploit identified at time of analysis, though a public technical writeup documents the flaw.
Reflected cross-site scripting in the WordPress plugin Tracking Code Manager (Data443) affects all versions up to and including 2.6.0, allowing unauthenticated attackers to inject arbitrary script into a page by supplying a payload in the TCM_ErrorMessages POST parameter, which the plugin reflects without sanitization or output escaping. Exploitation requires the attacker to lure a victim into submitting a crafted request or clicking a specially crafted link (UI:R), so each target must be individually induced to trigger the payload rather than being compromised by simply browsing a poisoned page. No confirmed active exploitation (CISA KEV) and no public exploit code were identified at time of analysis, and no EPSS score was supplied; the flaw is rated moderate (CVSS 6.1) with the main risk being session or credential theft against authenticated site users who are tricked into interacting with the crafted request.
Reflected cross-site scripting in the WordPress plugin Photo Gallery by 10Web - Mobile-Friendly Image Gallery (all versions through 1.8.42) allows unauthenticated attackers to inject arbitrary JavaScript into a page via the 'title' parameter, which the plugin fails to sanitize on input or escape on output. Exploitation is gated on user interaction: a victim - ideally an authenticated administrator - must be lured into clicking an attacker-crafted link, and because the flaw is reflected rather than stored, each target must be tricked individually, so this is assessed as a moderate rather than top-tier priority per our independent assessment (CVSS 3.1 base 6.1, AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:L/A:N). No public exploit code was identified at time of analysis and the issue is not confirmed actively exploited (CISA KEV); WordPress.org plugin tags indicate the flaw is addressed in version 1.8.43.
Cross-site scripting in ELADMIN 2.7 and earlier lets attackers inject script that executes in the browser of anyone who loads the affected components - the WebSocketServer code, SpringSecurityConfig, and the deploy.vue server-deployment view of the admin console - for a CVSS 3.1 base score of 6.1 (AV:N/AC:L/PR:L/UI:R/S:C/C:L/I:L/A:N). Exploitation requires victim interaction with the page carrying the payload, and because those screens normally sit inside the authenticated admin console the practical prerequisite may be an authenticated or attacker-controlled account; the published CVE vector lists PR:N while our assessment rates PR:L, so pre-authentication reachability is not confirmed and should be checked with the vendor. No public exploit code or CISA KEV activity was identified at time of analysis, and the impact is confined to the victim's browser session - script execution in the admin's origin, with no server-side code execution.
Denial of service in the @casl/ability npm package (6.7.5 through 7.0.1) stems from prototype pollution in the rulesToFields() helper, which expands attacker-influenced dotted condition keys through an internal setByPath utility and can overwrite inherited builtins such as Function.prototype.call, corrupting shared objects process-wide and crashing the Node.js process. Only applications that call rulesToFields() on rule conditions whose keys are under an attacker's write influence are affected, and the required access is authenticated (PR:L) with high attack complexity, so exposure is conditional rather than broad. The impact is availability only (no confidentiality or integrity loss), no public exploit code has been identified at time of analysis, and the flaw is not listed in CISA KEV, making this a moderate-severity issue rather than an urgent one.
Namespace-binding flaw in temporal-proxy 0.2.0 through 0.8.0 allows an authenticated caller authorized only for namespace B to obtain the decrypted plaintext of a payload that was encrypted for a different namespace A, provided payload encryption is enabled and the single proxy holds keys for more than one namespace. The proxy selects the decryption key from an identifier embedded in the payload rather than from the namespace of the request, and the ciphertext carries no namespace, so on 0.8.0 the caller can submit the payload to the codec server's /decode endpoint, while on every affected version a caller who can also write into namespace B on the Temporal server can place the payload there and read it back through the proxy. Impact is limited to confidentiality - payloads cannot be modified, the flaw grants no access to ciphertexts the attacker could not already obtain, and no public exploit was identified at time of analysis; severity is Medium (CVSS 4.0 base 6.0; independent assessment CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:N/A:N).
Sensitive information disclosure in Lenovo Smart Camera C33 - and the related C33 Pro, C35, RQ3, XW1, X6T, X6Q and Smart Network PTZ X5T/X5Q/X5E models - allows an attacker who is already positioned on the same adjacent network segment to read confidential data from the device without authenticating or relying on user interaction. The root cause class (CWE-1393, use of default password) points to weak or default credentials or secrets as the disclosure vector, and the CVSS v4.0 vector carries an AT:P precondition, meaning specific device state or configuration conditions must also hold, so this is not a direct internet-facing exploit. Exploitation is further limited because these cameras were distributed exclusively in the Chinese market; no public exploit code has been identified at time of analysis, and the independently assessed severity is a moderate, context-dependent risk rather than a top-tier emergency.
Improper validation of the Cato Point of Presence (PoP) server certificate during the Cato Client tunnel handshake (CWE-295) lets an attacker who already sits on the same network segment impersonate a PoP and read or rewrite tunneled traffic. The flaw is unauthenticated (CVSS vector PR:N) and needs no user interaction, but exploitation requires a man-in-the-middle position on the adjacent/local network (AV:A with attack requirement AT:P per the vendor vector, assessed as AC:H), so it is not internet-remote - rogue access points, ARP spoofing, or hostile hotspots are the realistic vectors, not default corporate deployments. Vendor and independent scoring put this at moderate severity (vendor CVSS 6.0; independent assessment CVSS:3.1/AV:A/AC:H/PR:N/UI:N/S:U/C:L/I:H/A:N, integrity impact highest), and no public exploit code was identified at time of analysis; it is a data-interception and traffic-tampering risk rather than a high-priority emergency.
Stack-based buffer overflow in HPE NonStop TELSERV (T9553L01^BAJ and earlier) allows unauthenticated remote attackers who can reach the Telnet server subsystem to corrupt memory and crash the service, producing a denial-of-service condition. The vulnerability is rated CVSS 5.9 (AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H): exploitation is network-reachable and requires no credentials, but the High attack complexity means the attacker must satisfy non-trivial conditions such as a specific memory state to reliably trigger the overflow, and the impact is confined to availability with no confidentiality or integrity effect. There is no public exploit identified at time of analysis, no CISA KEV listing, and NonStop deployments are typically confined to segmented back-office or payment networks rather than being internet-facing, making this a moderate-severity, limited-impact issue rather than a top-tier priority.
Canonical authd before 0.7.1 can silently rebind a recycled identity-provider email address to an existing local account record, so a newly created IdP account inherits the previous user's local UID and home directory - including SSH keys and cloud API tokens - on its first login to the authd-managed host. Exploitation is highly conditional rather than a mass-exploitation priority: it requires an OIDC broker deployment that maps the IdP email address to the local username, an IdP-side reassignment of that email to a new account after the original IdP account is removed, and an administrator who has not locked or deleted the departed user's local authd account; the flaw is local to the managed machine (CVSS:4.0 base 5.9; independently assessed as CVSS:3.1/AV:L/AC:H/PR:N/UI:N/S:U/C:H/I:L/A:N). No public exploit code was identified at time of analysis and the issue is not in CISA KEV, but the documented mitigation - locking or deleting the previous user's local authd account when the corresponding IdP account is removed - fully prevents it.
Local unprivileged users on SUSE Linux Enterprise Server 16.1, SLES for NVIDIA 16.1, openSUSE Leap 16.1 and openSUSE Tumbleweed can read a PIN or key secret that SUSE tooling renders as a QR code, because the secret is passed to qrencode as a command-line argument and is therefore exposed in /proc/<pid>/cmdline while the process runs. Exploitation requires the attacker to already hold a local low-privileged account on the same host (PR:L) and to win a short timing race against the brief qrencode invocation (AC:H), so this is a confidentiality-only credential-disclosure issue of modest severity with little realistic exposure on single-user or tightly controlled systems. There is no public exploit identified at time of analysis, and no vendor-released fix version has been confirmed in the available data, though SUSE tracks the issue through its Bugzilla and CVE pages.
Improper TLS certificate validation in Lenovo Smart Camera firmware lets a network-adjacent attacker who has already established a man-in-the-middle position on the same local or Wi-Fi segment intercept the camera's network communications and read the traffic it protects. Unauthenticated attackers are sufficient (PR:N), and only the victim's ordinary camera or companion-app activity is required for capture (passive user interaction), with high confidentiality impact but no availability impact. Affected devices include the Smart Camera C33, C33 Pro, C35, RQ3, XW1, X6T, X6Q, C5X and X9T plus the Smart Network PTZ X5T, X5Q and X5E, all sold exclusively in the Chinese market, which sharply narrows the realistic target population. No public exploit identified at time of analysis and no CISA KEV listing; at CVSS 4.0 base 5.9 this is a medium-severity, low real-world priority issue rather than an urgent one, consistent with the requirement to first obtain an adjacent MITM position and with the absence of internet-facing exposure.
A path traversal in OPNsense's OpenVPN Client Specific Override (CSO) handling allows an authenticated user holding CSO privileges to store a traversal sequence such as '../' in the common_name field, which openvpn_csc_conf_write() concatenates into the target path /var/etc/openvpn-csc/<vpnid>/<common_name> without normalization or a basename check; when a matching OpenVPN client-connect or authentication event is later processed by client_connect.php or user_pass_verify.php, a generated CSO configuration file is written outside the intended directory. Exploitation is constrained: the attacker must first have operator-level CSO access and must cause a matching common name or username to be processed, and the written content is generated OpenVPN configuration rather than arbitrary attacker-controlled bytes, which limits achievable impact - there is no public exploit identified at time of analysis. The issue is rated medium severity (vendor CVSS 5.9, independent assessment CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:L) and is fixed in OPNsense 26.1.11.
Insufficient session termination in the Apache Airflow Keycloak auth manager (apache-airflow-providers-keycloak before 0.11.1) leaves a user's session valid after logout, so anyone who later gains access to the same browser - for example on a shared, kiosk, or unattended workstation - can continue acting as that user with the victim's full privileges. The logout route does revoke the presented Airflow JWT, but KeycloakJWTMiddleware re-issues a fresh, unrevoked token in the same HTTP response, so the session never actually ends. The flaw is locally exploitable (CVSS:3.1/AV:L/AC:L/PR:N/UI:N, C:H/I:H/A:N, needing only subsequent local/physical access to the victim's browser), affects only deployments configured with the Keycloak auth manager, has no public exploit identified at time of analysis, and is fixed in apache-airflow-providers-keycloak 0.11.1.
Local information disclosure and symlink-following file overwrite in AsyncHttpClient versions 2.0.0 through 2.16.0 and 3.x before 3.0.12 let an authenticated local user on a shared multi-user host read resumable download URLs - including embedded credentials and query tokens - from a predictably named store at $java.io.tmpdir/ahc/ResumableAsyncHandler.properties, or plant a symbolic link that the client follows when saving state, truncating and overwriting any file writable by the client process. Only applications that decorate another AsyncHandler with ResumableAsyncHandler backed by PropertiesBasedResumableProcessor are exposed; the no-argument constructors use a no-op processor and are unaffected. Exploitation requires local access, control of the shared $java.io.tmpdir/ahc directory, and timing against a state save; no public exploit code was identified at time of analysis and there is no KEV confirmation of active exploitation. The issue is fixed in 2.16.1 and 3.0.12.
Out-of-bounds memory write in HPE NonStop RDMA Services NSK can corrupt memory, and the 5.8 CVSS score reflects a local attack path that also requires high privileges, user interaction, and high attack complexity. Only HPE NonStop systems using the RDMA Services NSK component are affected based on the supplied CPE, and no public exploit code or CISA KEV entry is present in the data. Realistic exploitation is limited to a privileged local user who can induce a separate user to interact under race/condition-dependent circumstances, making this a moderate-severity, low-priority issue in practice.
Stored cross-site scripting in Lychee photo-management software (all versions prior to 7.6.0) allows an authenticated user holding album-edit permissions to embed a javascript: URI inside an album or photo Markdown description, which is then converted to HTML server-side and rendered client-side via v-html in the gallery components. Because the payload is stored in album/photo metadata, any visitor who clicks the rendered link - including an unauthenticated visitor to a public album - executes attacker-controlled JavaScript in the Lychee origin, which can expose session data or let the attacker issue authenticated requests in the victim's context; exploitation requires the victim's click and therefore is not triggered by passive page views. The issue is fixed in Lychee 7.6.0 and no public exploit was identified at time of analysis.
An out-of-bounds memory read in HPE NonStop EXPAND can be triggered by an authenticated, highly privileged local operator on the NonStop platform, potentially leaking memory contents or destabilizing the EXPAND networking subsystem. Exploitation requires local access (AV:L), pre-existing high privileges (PR:H), and a hard-to-control memory or timing condition (AC:H), so it is neither remotely nor anonymously reachable; no user interaction is required. No public exploit code has been identified at time of analysis, and the vendor security bulletin (hpesbns05148en_us) is the authoritative source for affected releases and fixes.
An out-of-bounds read in zziplib through 0.13.80 allows attackers to crash applications that open a specially crafted ZIP archive via zzip_dir_open() when the library is compiled in mmap mode. The vulnerability triggers a 28-byte ZIP64 extra field over-read when an entry name ends exactly at the end of the mapped central directory. Exploitation requires a local attack vector, high complexity, and user interaction to open the crafted archive; there is no confidentiality or integrity impact, only availability loss (crash). No public exploit identified at time of analysis.
Plaintext credential disclosure in Eclipse Jetty occurs when a SessionAuthentication object holding a credential that contains a plaintext password is serialized into a persisted session data store, causing the password to be written to database, file, or cache storage in cleartext (CWE-312). The weakness is only exploitable in a specific deployment: session persistence must be enabled via a SessionDataStore, authentication must produce credentials containing a plaintext password, and the attacker must already hold sufficient privileges to read the persisted session data - database access, filesystem or backup access, or cache access. No public exploit identified at time of analysis, and CISA KEV does not list this issue; the independent assessment rates it a genuine but modest-priority flaw (CVSS 3.1 AV:L/AC:H/PR:H/UI:N/S:U/C:H/I:N/A:N, vendor CVSS 4.0 5.6) with confidentiality-only impact and no integrity or availability effect.
Intent redirection and improper access control in the AgentWeb Android WebView library (agentweb-core through 5.1.6-androidx) lets attacker-influenced web content loaded inside a host app's WebView launch that host app's own non-exported components and coerce URI permission grants for the host's ContentProviders by supplying crafted intent:// URLs that DefaultWebClient parses and fires under the host Activity's identity. The vulnerability is exploitable unauthenticated and remotely with no user interaction beyond the app rendering untrusted content (assessed CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:L/A:N), although real impact is bounded by the host app's own component and provider surface and by the host exposing sensitive non-exported components or providers; components belonging to other applications remain gated by the OS exported check. No public exploit identified at time of analysis, and the project has released a fixed version, 5.1.7-androidx, via commit 9efb1519.
Local privilege escalation in FreeBSD's sound(4) driver stems from a time-of-check/time-of-use race in audio-buffer resizing: a local, authenticated user with sound-device access can race an mmap(2) call against a buffer reallocation to cause a kernel use-after-free. Affected releases are FreeBSD 15.1-RELEASE before p5, 14.4-RELEASE before p11, and 14.5-RELEASE before p2; the independent assessment rates exploitation as local, authenticated, high-complexity, and requiring a non-deterministic timing win, with high confidentiality, integrity, and availability impact if successful. EPSS is 0.11% (1st percentile) and no public exploit identified at time of analysis, making this a genuine but lower-tier priority compared with remotely exploitable or actively exploited issues.
Arbitrary code execution is possible on a broad set of Lenovo consumer tablets - including the Idea Tab Pro TB373FU, Idea Tab / Lenovo Tab K11 Gen 2 TB336FU and TB336ZU, Legion Tab TB321FU, Tab M11 TB330FU and TB330XU, Tab Extreme TB570FU and TB570ZU, and Tab K11 TB330FUP, with additional models enumerated in vendor tagging (Tab K11 Plus, Tab K9, Tab One, Tab P12, Tab Plus, Yoga Tab, Yoga Tab Plus, and the TB311FU/TB311XU folio and clear-case variants) - by an attacker who has physical possession of the tablet while it is already unlocked. The CVSS 4.0 base score is 5.4 (MEDIUM), with vector AV:P/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N, meaning exploitation requires the attacker to be physically at the device (AV:P), no prior privileges are needed (PR:N), no victim interaction is required beyond the device already being unlocked (UI:N), and attack requirements are present (AT:P) in the form of physical custody plus an unlocked state; the flaw affects a device's own confidentiality, integrity and availability but does not propagate to subsequent systems. No CISA KEV listing is associated with this CVE and no public exploit code has been identified at time of analysis, so real-world risk is bounded by the physical and unlocked-device preconditions - a locked, powered-off, or patched unit is not exploitable via this issue.
Stored cross-site scripting in OPNsense core before 26.1.11 allows a user holding Firewall Categories or Firewall Alias Edit privileges to plant script that executes in the browser session of an administrator or other operator who opens the Firewall Rules or NAT page. The injected code runs with the victim's context, can read CSRF tokens, and can therefore perform authenticated actions on the firewall as that victim. Exploitation is not anonymous - it requires the attacker to already possess category or alias editing rights (PR:L, CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:L/I:L/A:N, base 5.4), and the malicious category must be attached to a rule or the alias referenced as a rule source/destination and saved so the tooltip renders; no public exploit code was identified at time of analysis and the issue is not confirmed as actively exploited in CISA KEV. Fixed in OPNsense 26.1.11.
Stored cross-site scripting (XSS) in JetBrains TeamCity's classic build log allows an authenticated user with permission to inject content into build output to execute arbitrary JavaScript in the browser of another user who views the affected log. The flaw affects TeamCity versions before 2026.2, 2026.1.4, and 2025.11.8, and exploitation requires both authentication and victim interaction, making it a moderate-severity risk (vendor CVSS 6.7). No public exploit code has been identified at time of analysis.
Stored cross-site scripting (XSS) in JetBrains TeamCity versions before 2026.2, 2026.1.4, and 2025.11.8 allows an authenticated low-privilege attacker to inject malicious scripts into the classic project overview page, which execute in the context of any user-including higher-privileged administrators-who navigates to that view. The vulnerability requires the attacker to have a valid TeamCity account with rights to store content that renders in the classic overview, and a privileged victim must actively open that legacy view; no public exploit code has been identified at time of analysis. The independent assessment rates the overall risk as conditional rather than critical, noting that instances not exposing the classic overview to low-privilege contributors are not realistically exploitable.