Elasticsearch's machine learning API contains an unbounded resource allocation flaw that allows an authenticated user with elevated privileges to crash an affected node by submitting a specially crafted ML request over the network. The absence of throttling or quota enforcement on ML request memory consumption (CWE-770) means the node's JVM heap can be exhausted, rendering it unavailable without any confidentiality or integrity exposure. No public exploit code has been identified at the time of analysis, and this vulnerability does not appear in the CISA KEV catalog.
Prototype pollution in @hey-api/openapi-ts affects all versions through 0.97.2, allowing remote unauthenticated attackers to substitute the prototype chain of the returned params slot object by passing a crafted key such as '$query___proto__' through any application that forwards user-supplied parameters to a generated SDK method. The flaw resides in a runtime template (dist/clients/core/params.ts) that is copied verbatim into every generated SDK, meaning every downstream npm package regenerated from this tool carries the vulnerable code - confirmed affected consumers include @opencode-ai/sdk and @trigger.dev/sdk. A functional proof-of-concept is publicly available; exploitation is not confirmed as actively exploited and is absent from CISA KEV.
Stored XSS in MyComplianceOffice MCO version 25.3.3.1 allows a privileged attacker with logo-upload rights to plant malicious JavaScript inside a crafted SVG file that executes in any user's browser when the application logo is rendered. Reported by CERT-PL, vendor contact was unsuccessful, leaving no official patch or advisory in place. No public exploit code has been identified at time of analysis, though the SVG-based XSS technique is well-documented and trivially replicable against unpatched instances.
Race condition in the Linux kernel signal subsystem allows a local low-privileged user to trigger a kernel warning and potential system instability by exploiting a timing window between SIGSTOP delivery and a concurrent execve() call in a multi-threaded process. The calling thread in zap_other_threads() retains stale JOBCTL_STOP_PENDING and JOBCTL_STOP_CONSUME flags after the thread group is torn down, causing task_participate_group_stop() to decrement an already-zero group_stop_count and fire a kernel WARN_ON. No public exploit has been identified and EPSS sits at 0.16% (6th percentile), consistent with a difficult-to-reliably-trigger race.
Uncontrolled resource consumption in ImageMagick (all versions prior to 6.9.13-51 and 7.1.2-26) allows a denial-of-service condition by supplying invalid arguments to the connected-components processing option, which causes an infinite loop that exhausts CPU resources indefinitely. The vulnerability requires local access, user interaction, and high attack complexity, confining realistic risk to environments where untrusted input can influence ImageMagick invocations - such as automated image-processing pipelines or web applications that expose connected-components functionality. Vendor-released patches exist for both the 6.x and 7.x branches; no public exploit code and no active exploitation (CISA KEV) have been identified at time of analysis.
Unbounded mount-time loop in FatFs prior to R0.16 allows physical attackers to cause a denial of service by presenting a crafted GPT disk image with an arbitrarily large GPTH_PtNum partition count field. The flaw is present only in builds where the FF_LBA64 = 1 compile-time flag is set, enabling 64-bit LBA and GPT scanning support. A public proof-of-concept is available from runZero; no confirmed active exploitation (CISA KEV) has been observed, and SSVC categorizes technical impact as partial, limited to availability.
Divide-by-zero in FatFs R0.16 and earlier's exFAT sync logic crashes the filesystem when crafted volume metadata causes the expression `n_fatent - 2` to evaluate to zero during write/sync operations, resulting in a hard fault or denial of service on the affected embedded device. The elm-chan FatFs library is widely bundled into microcontroller SDKs and IoT firmware, meaning the vulnerable code exists across a broad range of downstream products compiled with exFAT support. A proof-of-concept exists per SSVC assessment; no confirmed active exploitation (CISA KEV) has been recorded, and the physical attack vector limits mass exploitation, though network-accessible firmware update pipelines can extend effective reach.
Uninitialized cluster exposure in FatFs R0.16 and earlier allows recovery of residual disk data when f_lseek() extends a file beyond its current EOF without zeroing newly allocated FAT clusters. Any application or device reading the extended region receives raw, previously written sector content instead of zeroed bytes, leaking sensitive residual storage data such as deleted file fragments or cryptographic material. A proof-of-concept exploit exists per SSVC and runZero's GitHub repository; no active exploitation has been confirmed in CISA KEV.
Plugin directory guard bypass in HashiCorp Vault and Vault Enterprise allows a highly privileged, authenticated attacker to read files outside the intended audit plugin directory via path traversal. Exploitation requires network access, high-privilege credentials, and the use of the legacy file audit path option - a non-default configuration. Fixed versions are 2.0.1, 1.21.6, 1.20.11, and 1.19.17; no public exploit or active exploitation (CISA KEV) has been identified at time of analysis.
Sensitive HTTP request header values are written into Kibana application logs when the optional APM (Application Performance Monitoring) instrumentation feature is enabled, exposing credentials or tokens to anyone with operator-level log access. This affects multiple Kibana release branches (8.18.x, 8.19.x, 9.0.x, and 9.1.x) and is classified as information disclosure under CWE-532. No public exploit or active exploitation has been identified at time of analysis; the exposure is passive and gated behind both a non-default configuration and existing operator privileges.
Kadence Blocks WordPress plugin versions up to and including 3.7.7 allows authenticated Contributor-level users to read or delete optimizer analysis records belonging to posts they do not own, by exploiting a parameter mismatch in four REST API endpoints of the Optimize_Rest_Controller. The authorization check binds to a user-supplied post_id while the storage layer retrieves records by sha256 of a separately supplied, attacker-controlled post_path, with no enforcement that these two parameters refer to the same post. No public exploit code or CISA KEV listing exists at time of analysis; however, the low barrier to exploitation - any registered Contributor can leverage this - makes it relevant for multi-author WordPress sites. EPSS data was not provided in the input.
Authorization bypass in the Kadence Blocks WordPress plugin (all versions through 3.7.7) allows authenticated contributors to create arbitrary Media Library attachments by fetching remote images via internal WordPress functions, bypassing the upload_files capability boundary. The flaw resides in class-kadence-blocks-prebuilt-library.php at multiple call sites where wp_upload_bits() and wp_insert_attachment() are invoked without verifying the requesting user's authorization. No public exploit has been identified at time of analysis, and the vulnerability is not listed in the CISA KEV catalog.
Unauthorized private content disclosure in the Slim SEO WordPress plugin (all versions through 4.9.8) allows authenticated Contributors to read protected content they do not own via the `/wp-json/slim-seo/meta-tags/ai` REST API endpoint. The endpoint's broken object-level authorization - checking only the generic `edit_posts` capability rather than per-post read access - lets any Contributor supply an arbitrary post ID and receive an AI-generated summary of that post's raw content, including private, draft, pending, future, and password-protected posts authored by other users. No public exploit or CISA KEV listing is identified at time of analysis; real-world risk is constrained by the requirement for a pre-existing Contributor-level account.
Arbitrary group deletion in the JoomSport WordPress plugin (versions ≤5.7.8) is achievable by any authenticated subscriber-level user due to a missing capability check in the joomsport_season_groupdel() AJAX handler. The handler validates a WordPress nonce - preventing CSRF - but never calls current_user_can() or equivalent, allowing any logged-in user to supply attacker-controlled group IDs to an unguarded DELETE query. No public exploit has been identified at time of analysis and this vulnerability is not listed in CISA KEV; real-world risk is bounded by the requirement for a valid WordPress account and the limited scope of impact (group record deletion only).
Cross-tenant information disclosure in Foreman (packaged as Red Hat Satellite 6) allows an authenticated user holding host-edit permissions to retrieve sensitive infrastructure metadata from organizations and locations they are not authorized to access. The root cause is the taxonomy_scope controller method accepting organization and location IDs from nested request parameters without validating them against the caller's authorized tenancy scope, implementing a textbook CWE-639 authorization bypass through user-controlled key. No active exploitation has been confirmed (not listed in CISA KEV), and no public exploit code is identified at time of analysis; however, the low attack complexity and lack of user interaction make this straightforward to exploit for any low-privileged authenticated user in multi-tenant Satellite deployments.
Authorization bypass in the Motors - Car Dealership & Classified Listings Plugin for WordPress allows authenticated subscriber-level users to arbitrarily mark any other user's vehicle listing as sold by replaying a harvested nonce against a victim's post ID. All plugin versions up to and including 1.4.111 are affected. An attacker exploiting this flaw can silently deface competitor or victim listings with a site-wide 'Sold' badge while also stripping the `special_car` featured post meta, causing business harm on dealership or classifieds sites. No public exploit has been identified at time of analysis and this CVE is not listed in the CISA KEV catalog.
Sensitive PII exposure in the Appointment Booking Calendar WordPress plugin (versions up to and including 1.4.02) allows any authenticated user with contributor-level access to retrieve customer booking records - including names, email addresses, phone numbers, and appointment comments - via the unprotected `cpabc_appointments_filter_list` AJAX action. The root cause is CWE-862 (Missing Authorization): the endpoint performs no capability check before returning data, meaning the attacker's only prerequisite is a valid WordPress login at contributor tier or above. No public exploit code or active exploitation (CISA KEV) has been identified at time of analysis, but a source-level fix appears committed in the plugin's SVN repository (changeset 3581633).
Insecure Direct Object Reference in the Qi Blocks WordPress plugin (all versions ≤ 1.4.9) allows any authenticated user with Author-level access to overwrite the stored global styles of arbitrary posts, templates, or widgets they do not own by manipulating the unvalidated 'page_id' parameter. Critically, the reserved 'template' and 'widget' page_id values expose site-wide surfaces, enabling an Author to deface, hide content on, or visually degrade any page across the entire WordPress installation. The permission_callback validates only generic edit_posts and publish_posts capabilities rather than post ownership, meaning the built-in Author role is sufficient for exploitation. No active exploitation has been confirmed (not listed in CISA KEV), and no public exploit code has been identified at time of analysis.
UI spoofing in Google Chrome's Skia graphics engine (versions prior to 150.0.7871.46) allows a remote attacker who has already compromised the renderer process to mislead users through a crafted HTML page. The attack requires renderer process compromise as a prerequisite, making this a chained exploit component rather than a standalone threat. EPSS at 0.18% (8th percentile) and absence from CISA KEV confirm no known active exploitation; Chromium's own team rated this Low severity.
Cross-origin data leakage in Google Chrome's ANGLE graphics subsystem (all versions prior to 150.0.7871.46) enables remote attackers to read sensitive data from other browser origins by luring a victim to a crafted HTML page. The flaw stems from uninitialized memory use in ANGLE, Chrome's GPU abstraction layer, which may expose stale cross-origin pixel buffers or texture memory to attacker-controlled JavaScript. No public exploit has been identified and CISA's SSVC framework rates exploitation as none with a non-automatable attack path, indicating limited immediate real-world threat despite the network-accessible vector.
Authorization bypass in Wagtail CMS allows authenticated low-privilege users to create translations for pages beyond their permitted scope. Versions prior to 7.0.8, 7.3.3, and 7.4.2 fail to enforce page-level permission checks when a user submits a translation request, exposing content of restricted pages to anyone holding the 'Can submit translation' permission. No public exploit code or active exploitation (CISA KEV) has been identified, but the low attack complexity and network-accessible vector make this straightforward to abuse for information disclosure.
Wagtail's Documents and Images chooser endpoint leaks asset metadata - including filenames, display names, and URLs - to authenticated admin users who have been explicitly denied collection-level choose permissions. Versions prior to 7.0.8, 7.3.3, and 7.4.2 are affected across the 7.0, 7.3, and 7.4 release branches. An attacker with a valid Wagtail admin account can enumerate restricted media assets across collections beyond their authorized scope, bypassing the collection permission model. No public exploit or active exploitation (CISA KEV) has been identified at time of analysis.
Cross-Site Request Forgery in GiveWP - Donation Plugin and Fundraising Platform (versions up to and including 4.15.3) allows unauthenticated remote attackers to disable donation email notifications by tricking a logged-in administrator into clicking a crafted link. The root cause is absent nonce validation in the give_set_notification_status_handler() AJAX handler, permitting forged state-changing requests. No public exploit code has been identified and this CVE is not listed in the CISA KEV catalog; however, the integrity impact - silently disabling donation notifications - could cause operational blind spots for nonprofit or fundraising site operators.
The genucenter web interface by genua exposes SNMP authentication and encryption keys in cleartext HTTP responses to any user holding the 'Service' or 'Admin' role, enabling credential theft without any exploitation complexity beyond normal authenticated use. Affected versions are all releases prior to 8.0p11; the fix was disclosed by sba-research via a coordinated advisory in April 2026. No public exploit code or active exploitation has been identified at time of analysis, though the disclosure of SNMP v3 keys - which govern both authentication integrity and traffic encryption - represents a meaningful lateral movement risk if those keys are reused across network devices.
SSRF policy bypass in jshookmcp 0.3.1 allows an authenticated MCP client with network domain access to probe internal RFC 1918 and reserved addresses that are explicitly blocked by all other network tools on the same server. The `network_icmp_probe` and `network_traceroute` handlers call `resolveHostname` directly without invoking the central `resolveAuthorizedTransportTarget` guard, creating an inconsistent enforcement boundary. No CISA KEV listing exists, but proof-of-concept test code demonstrating the bypass via the `handleCallTool` dispatch path is included in the GitHub advisory (GHSA-c5r6-m4mr-8q5j), confirming exploitability without external traffic.
Missing authorization in the ThumbPress WordPress plugin (Codexpert Inc) through version 6.3.2 allows low-privileged authenticated users to invoke restricted plugin functionality, resulting in limited availability impact against thumbnail or image-size management operations. The CVSS vector (PR:L, A:L) confirms exploitation requires an authenticated session but no elevated role, and produces no confidentiality or integrity loss - only partial disruption. No public exploit code has been identified at time of analysis, and the vulnerability is not listed in the CISA KEV catalog.
Administrative access to the Stormshield Network Security captive-admin portal is bypassed when an attacker presents a revoked client certificate, because the portal does not enforce certificate revocation checks during TLS mutual authentication. Affected versions span three branches: 4.3.0-4.3.41, 4.4.0-4.8.15, and 5.0.2 EA-5.0.5. No public exploit has been identified at time of analysis and the vulnerability is absent from CISA KEV, but the threat actor population is meaningfully constrained - exploitation requires prior possession of a certificate that was specifically issued for this portal and then revoked.
Code injection in the Wikimedia Foundation Timeline extension (EasyTimeline) allows authenticated wiki editors to execute arbitrary server-side code via crafted timeline syntax processed by the Perl-based EasyTimeline.Pl script and its PHP integration layer Timeline.Php. The vulnerability stems from CWE-94 improper code generation control, where user-supplied timeline markup is passed to the backend Perl interpreter without sufficient sanitization, enabling remote code execution on the MediaWiki server. No public exploit code has been identified at time of analysis, but the RCE tag and network-accessible attack vector make this a high-priority patching target for any publicly editable or semi-public MediaWiki instance.
Cookie tossing in AsyncHttpClient (AHC) library allows a malicious HTTP server to plant cookies scoped to an unrelated trusted domain, affecting versions 2.0.0-2.15.x and 3.0.0.Beta1-3.0.10. ThreadSafeCookieStore accepts and stores a cookie's Domain attribute value without verifying that the responding host is authorised to set cookies for that domain, so any server the client contacts can inject a cookie that AHC will automatically forward to the targeted domain on subsequent requests. Java applications sharing a single AHC instance - and therefore its default shared CookieStore - across calls to both attacker-influenced and trusted hosts are the primary attack surface; no public exploit has been identified at time of analysis, and vendor-released fixes are available in 2.16.0 and 3.0.11.
UltraVNC Repeater through 1.8.2.2 harbors a latent off-by-one stack buffer boundary condition in its HTTP Basic authentication Base64 decoder, where a strict greater-than comparison at `repeater/webgui/webutils.c:817` fails to block an input whose length exactly equals the 1024-byte output buffer. Under current code, the outer HTTP request parser incidentally caps Authorization header length before the defect can produce an out-of-bounds write, making this vulnerability practically unexploitable in its present form - but the flaw is real and would become a one-byte stack write if upstream buffering constraints change. No public exploit code exists and the vulnerability is not listed in CISA KEV; this is a latent memory-safety defect requiring patch application as hygiene rather than urgent incident response.
Wagtail CMS versions prior to 7.0.8, 7.3.3, and 7.4.2 expose a resource exhaustion vector where authenticated admin users can submit purposefully crafted image filter specification strings that trigger disproportionately expensive rendition processing on the server, resulting in service degradation for all site users. The vulnerability is constrained to admin-level authenticated users and cannot be triggered by ordinary unauthenticated site visitors, materially limiting real-world attack surface. No public exploit code and no CISA KEV listing have been identified at time of analysis, placing this in the low-priority tier for most deployments.
Zip slip (relative path traversal) in ClearML's StorageManager._extract_to_cache() allows a network-positioned attacker to write arbitrary files to the host filesystem, with confirmed RCE potential via cron job injection, SSH authorized_keys overwrite, or web shell deployment. Affected are all versions up to and including 1.16.5; the official CVSS score of 2.4 with I:L severely understates the real-world impact given the documented RCE paths. No KEV listing and no EPSS data were provided, but the presence of a public bounty report and a confirmed fix commit indicates the issue is verified. The fix is available in version 2.1.6.
Resource exhaustion in Wasmtime's native WASIp1 implementation allows low-privileged WebAssembly guests to exhaust host-level file descriptors and OS resources by repeatedly invoking fd_renumber in a loop. The affected versions span four distinct release branches - all pre-24.0.10, 25.x-35.x, 37.x-44.x, and 45.0.0-45.0.1 - but only runtimes that both expose fd_renumber and grant guests the ability to open files are vulnerable. No public exploit code exists and the issue is not listed in CISA KEV; however, the attack is mechanically straightforward once the conditions are met, making patching the primary defense.
Unvalidated bearer realm URL handling in oras-go v2 ≤ 2.6.0 enables two distinct attack primitives against users who run oras operations against a malicious, compromised, or man-in-the-middle'd registry: server-side request forgery (SSRF) to internal network endpoints including cloud instance metadata services, and TLS downgrade that exposes user credentials in plaintext. The OCI distribution spec legitimately allows cross-host realm references for split-auth deployments (e.g., Docker Hub's auth.docker.io pattern), but oras-go failed to block private IP literals, loopback addresses, and scheme downgrades from https to http - patterns that are never legitimate under any valid registry trust model. No active exploitation has been confirmed (not in CISA KEV), and a patch is available in v2.6.1.
Insufficient permission enforcement in MediaWiki's user query API endpoints allows low-privileged authenticated users to access sensitive user information - such as group memberships or account details - that permission controls should restrict. The vulnerability spans four source files: ApiQueryAllUsers.php, ApiQueryUsers.php, PermissionManager.php, and UserGroupManager.php, indicating that the flaw lies in how permission checks are applied (or bypassed) when the API serializes user data. No public exploit code exists and the CVSS 4.0 score of 2.1 reflects the constrained real-world exploitability. No confirmed active exploitation (CISA KEV) has been reported.
Cross-site scripting in Wikimedia Foundation MediaWiki and the CentralAuth extension allows authenticated low-privileged users to inject malicious scripts into rendered wiki pages or API output through unsanitized input in API formatting, ResourceLoader module handling, page display hooks, and permission change log formatting. Affected versions span all releases of both products prior to the 1.46.0, 1.45.4, 1.44.6, and 1.43.9 patch series. No public exploit code or CISA KEV listing has been identified at time of analysis, though the five affected PHP files suggest a broad attack surface across core MediaWiki subsystems.
Cross-site scripting vulnerabilities across multiple log formatter and language components in MediaWiki allow authenticated low-privilege users to inject malicious JavaScript that executes in the browser of any user who views affected log pages or the SpecialVersion special page. The vulnerability spans seven PHP files covering block, patrol, rename-user, and tag log formatters, the base LogFormatter, the Language class, and SpecialVersion - indicating widespread unsanitized output rendering in administrative and logging interfaces. No public exploit code has been identified and the vulnerability is not listed in CISA KEV, but the breadth of affected components and the multiple concurrent fix branches (1.43.9, 1.44.6, 1.45.4, 1.46.0) signal a significant audit-driven remediation effort.
Information disclosure in MediaWiki's core parser component (includes/Parser/Parser.php) allows authenticated low-privilege users, under conditions requiring user interaction, to obtain sensitive information not intended for their access level. All MediaWiki branches are affected up to versions 1.43.9, 1.44.6, 1.45.4, and 1.46.0, covering a broad installation base including self-hosted wikis worldwide. No public exploit code exists and no active exploitation has been confirmed at time of analysis; however, the provided CVSS 4.0 vector contains a contradictory zero-impact scoring that warrants independent verification against the upstream advisory.
Improper input validation in Wikimedia Foundation's UrlShortener extension (includes/UrlShortenerUtils.php) exposes a potential information disclosure path accessible to low-privileged authenticated users over the network. The vulnerability is classified under CWE-20 and tagged as an information disclosure issue, though the provided CVSS 4.0 vector records zero impact across all dimensions - a notable contradiction with the 'Information Disclosure' tag that warrants independent verification. No public exploit code exists and no active exploitation has been confirmed.
Session fixation in the Wikimedia Foundation OAuth MediaWiki extension (src/Backend/MWOAuthServer.php) allows a network-accessible, low-privileged attacker to manipulate the OAuth authorization handshake so that a victim's authentication binds to an attacker-controlled session identifier. All four actively maintained release branches are affected through versions 1.46.0, 1.45.4, 1.44.6, and 1.43.9. No public exploit or CISA KEV listing exists at time of analysis; however, the 'Information Disclosure' tag applied to this CVE creates tension with the vendor-supplied CVSS 4.0 zero-impact scoring and warrants independent verification.
Cross-site scripting in the Wikimedia Foundation CheckUser extension (versions 1.46.0-rc.0 up to but not including 1.46.0) allows a high-privileged attacker to inject malicious scripts via the blockConnectedTempAccountsField Vue component used in the Temporary Accounts blocking workflow. Exploitation requires both elevated CheckUser-level privileges and active user interaction from another party, materially constraining real-world risk. No public exploit code exists and this CVE is not listed in CISA KEV, indicating no confirmed active exploitation at time of analysis.
Cross-site scripting in MediaWiki's Special:Block Vue.js component allows a high-privileged user - such as a wiki administrator - to inject malicious scripts that execute in the browsers of other users who subsequently view the affected block interface output. The vulnerability is isolated to the `SpecialBlock.Vue` frontend component, gating exploitation behind administrative access. No public exploit has been identified at time of analysis, and no patch version has been confirmed from available data.
Stored cross-site scripting in the Wikimedia Foundation Timeline extension exposes MediaWiki installations to session hijacking and credential theft. The flaw exists in Timeline.php and EasyTimeline.pl, where user-supplied timeline markup syntax is not properly neutralized before HTML output, allowing an authenticated wiki editor to inject arbitrary JavaScript executed in other users' browsers. No public exploit has been identified at time of analysis, and KEV listing is absent, but the network-accessible nature and low-privilege entry point make this a credible risk on publicly editable wikis.
Cross-site scripting in MediaWiki's ApiSandbox component (ApiSandboxLayout.js) allows an authenticated attacker to inject malicious JavaScript that executes in the browser of a victim who interacts with a crafted ApiSandbox page. Only MediaWiki 1.46.0-rc.0 is confirmed affected; the 1.46.0 stable release resolves the flaw, meaning real-world exposure is limited to deployments running the release candidate. No public exploit code or active exploitation has been identified at time of analysis.
Remote OS command injection in Network UPS Tools upsmon (versions 2.8.3-2.8.5) allows a network-positioned attacker - acting as a man-in-the-middle on the plaintext 3493/tcp NUT protocol connection, operating a compromised upsd server, or controlling a rogue UPS device - to execute arbitrary shell commands as the upsmon service account. The vulnerability exists in the NOTIFYCMD alarm handler, where server-supplied ups.alarm text is interpolated directly into a shell command string and executed via system(), enabling injection of shell metacharacters. No public exploit code or active exploitation (CISA KEV) has been identified at time of analysis, though the oss-security disclosure contains sufficient implementation detail to construct a working exploit.
Open redirect in Concourse's web login flow allows a remote unauthenticated attacker to craft a URL that silently redirects an authenticated user to an arbitrary external site upon completing the login process. Concourse versions prior to 8.2.3 are affected via the `/sky/login` endpoint's `redirect_uri` parameter. While no credentials are leaked server-side, the post-login redirect to an attacker-controlled site creates a high-fidelity phishing vector targeting CI/CD operators and developers - users who typically hold sensitive pipeline secrets and source-code access. No public exploit identified at time of analysis, though the advisory provides a fully working proof-of-concept URL.