Heap buffer overflow in Suricata 8.0.0 through 8.0.5 lets a user who can supply or modify a local detection rule crash the engine by loading a ruleset that pairs frame inspection without a content keyword with a transformed match that also lacks content. The flaw is reached during signature preparation in src/detect-engine-prefilter.c and fires even in config-test mode (suricata -T), so no live traffic is required, but it is strictly local, requires rule-loading privilege, and results only in a denial-of-service crash rather than code execution. This is a low-severity issue: the assessed CVSS vector is AV:L/PR:L with A:L impact only, there is no public exploit identified at time of analysis, and Suricata 8.0.6 plus the entire 7.x line are not affected.
OS command injection in the Spatie Scotty developer CLI (through v1.4.4) lets a malicious 'host' value in a Scotty @servers configuration directive break out of the local shell and execute arbitrary commands when a victim runs the Doctor command. Because the host string is interpolated unescaped into Process::fromShellCommandline, shell metacharacters such as ';' and '${IFS}' in a crafted host are interpreted by the local shell rather than passed literally to ssh. This is a local CLI tool, so exploitation requires the attacker to get malicious configuration in front of the victim and the victim to subsequently invoke Doctor (checkSshConnectivity or checkRemoteTools), and any injected commands run only with the victim's own user privileges; no public exploit code has been identified at time of analysis, and vendor CVSS 4.0 rates this 5.3 with current exploit maturity unproven (E:X).
Local information disclosure in cockpit-machines on Red Hat Enterprise Linux 9 and 10 allows an authenticated local user to read plaintext VM credentials - including passwords - that the component passes directly on process command-line arguments during VM creation or installation. Exploitation is timing-dependent: the secret is only readable while a provisioning operation is in flight, and it also requires that /proc process arguments remain visible to other users (i.e., the host is not hardened with a hidepid-style mount option), so risk concentrates on multi-user or shared virtualization hosts. It yields no code execution and no privilege escalation, but a disclosed password compromises access to the affected VM; no public exploit code has been identified at time of analysis and the flaw is not confirmed actively exploited (not in CISA KEV).
Rendering a crafted PDF through Poppler's Splash backend can pin a CPU core at 100% for an extended attacker-controlled duration. The flaw in SplashOutputDev::tilingPatternFill computes an excessive loop count from tiling-pattern geometry near the int32 limit, without allocating memory. This availability-only issue affects Red Hat Enterprise Linux 6, 7, 8, 9, and 10, and Red Hat Hardened Images, and exploitation requires a victim to open a malicious PDF (UI:R, AV:L, PR:N - no privileges required). No public exploit identified at time of analysis, and no confirmed active exploitation.
Unauthenticated information disclosure affects the Multicluster Observability Addon in Red Hat Advanced Cluster Management for Kubernetes, enabling attackers on the adjacent network to access a debug endpoint that leaks goroutine, heap, and command-line profiling data. The misconfiguration originates in the underlying addon-framework library, which exposes the endpoint without requiring credentials; exploitation only requires a basic encrypted (TLS) connection. No public exploit has been identified at time of analysis, and the vulnerability does not permit remote code execution, but the exposed profiling information could aid reconnaissance or further attacks.
IBM Controller 11.0.0 through 11.0.1 FP7 and 11.1.0 through 11.1.3 FP1 permit an authenticated low-privilege user to bypass file-size validation by tampering with client-side input, producing low integrity and availability impact without confidentiality exposure. The issue is rated CVSS 5.4 and assessed as a moderate, not a priority, issue; no public exploit identified at time of analysis. A vendor patch is available per IBM advisory, and exploitation requires a valid IBM Controller account with network access.
Stored cross-site scripting in the FileBird WordPress media library and file manager plugin (all versions up to and including 6.5.6) allows authenticated users with Author-level capabilities or higher to inject arbitrary JavaScript through an image's alt-text field, which is then rendered without adequate escaping in FileBird gallery and PhotoSwipe output. The payload persists in attachment metadata and executes in the browser of any user who later views an affected gallery page, including administrators, so a low-privileged content author can pivot into an administrative session context. The vendor score is moderate (5.4) and our assessment raises the scope to changed (S:C); no public exploit code has been identified at time of analysis and no EPSS or KEV signal was provided, so this should be handled as a routine patch-in-cycle item rather than an emergency.
Denial of service in IBM i 7.3, 7.4, 7.5, and 7.6 allows unauthenticated attackers on an adjacent network segment to disrupt the integrated FTP server by sending authentication commands that are not properly validated. Because the assessed vector is CVSS:3.1/AV:A/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L, the credential-free USER/PASS handshake is the likely trigger path, but realistic exposure is confined to LAN or same-broadcast-segment reachability rather than arbitrary internet access, and there is no confidentiality exposure or code execution evidenced. IBM has published a vendor fix, and no public exploit code or confirmed active exploitation was identified at time of analysis; this is a moderate-severity, low-real-world-priority issue.
Cross-site scripting in IBM Common Licensing Agent and ART versions 9.0, 9.0.0.1, and 9.0.0.2 allows a user who already holds a low-privilege account on the licensing Web UI (PR:L) to embed arbitrary JavaScript that executes in another user's trusted session when that victim loads the tainted page (UI:R), potentially disclosing credentials or other session data. The flaw is rated CVSS 3.1 5.4 (AV:N/AC:L/PR:L/UI:R/S:C/C:L/I:L/A:N) and carries CWE-79; exploitation is limited to authenticated attackers and depends on a victim viewing the affected content, so it is not remotely exploitable without credentials or user interaction. No public exploit code was identified at time of analysis and there is no confirmed active exploitation (not listed in CISA KEV); a patch is available from the vendor.
Stored cross-site scripting in the WP Recipe Maker plugin for WordPress affects all versions up to and including 10.8.1, allowing an authenticated attacker holding Contributor-level access or above to inject arbitrary JavaScript through the recipe 'notes' parameter, which is neither properly sanitized on input nor escaped on output. The injected script executes in the browser of any user - visitor, editor, or administrator - who subsequently views the affected recipe page, and per the CVSS vector (PR:L, UI:R) the attack requires both a valid low-privilege account and a victim's interaction with the poisoned page. There is no public exploit code identified at time of analysis and the issue is not confirmed actively exploited (CISA KEV), so exposure is confined to multi-author WordPress installations that grant Contributor accounts to untrusted users; the scope-change rating (S:C) reflects that successful script execution extends impact beyond the plugin itself into the surrounding site's user sessions.
Cross-site scripting in the Video API of Synology DiskStation Manager (DSM) allows a remote authenticated user to read or write a limited set of files in the context of a victim's browser session when the video player is launched. Affected builds are DSM before 7.2.1-69057-10, 7.2.2-72806-7, and 7.3.2-86009-2; exploitation requires a valid DSM account with Video API access plus victim interaction (launching the player), and impact is bounded to low-confidentiality and low-integrity file access rather than code execution. No public exploit code or confirmed active exploitation is identified in the available data, and the CVSS 5.4 / EPSS-consistent profile places this as routine-priority patching.
Cross-site scripting in the shared Biggop promotional library bundled with seven bdthemes Elementor plugins - Element Pack Addons for Elementor ≤8.7.14, Prime Slider ≤4.4.5, Ultimate Post Kit ≤4.2.0, Ultimate Store Kit ≤3.0.7, Pixel Gallery ≤2.1.14, Live Copy Paste ≤1.5.6, and Smart Admin Assistant ≤2.2.0 - allows arbitrary script injection through the unescaped 'display_id' value returned by the external Sigmative/Biggop API. Exploitation is not trivial: an attacker must first compromise or impersonate that third-party API server, and a victim (typically a logged-in administrator) must then load the plugin admin page that renders the tainted API response through the biggopti JavaScript. There is no confirmed active exploitation (not listed in CISA KEV) and no public exploit code identified at time of analysis; the CVSS 5.4 score reflects high attack complexity (AC:H) and the scope change rather than a directly reachable flaw in the plugin itself.
Unauthenticated remote attackers can degrade the email service of IBM Sterling Partner Engagement Manager (PEM) by flooding it with requests, exploiting improper control of interaction frequency (CWE-799) so the service is unable to keep up with legitimate mail processing. The affected releases are Essentials Edition 6.3.0.0 through 6.3.0.2 and 6.2.4.0 through 6.2.4.4, plus Standard Edition 6.2.4.0 through 6.2.4.4; per the assessed CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L, base 5.3 Medium), no authentication or user interaction is required against a default configuration when the email service is network-reachable. Impact is limited to partial availability loss (A:L) with no confidentiality or integrity effect and no scope change, so this is a service-degradation weakness rather than a full application outage; no public exploit code has been identified and the flaw is not listed in CISA KEV, so there is no public exploit identified at time of analysis.
IBM Controller releases 11.0.0 through 11.0.1 FP7 and 11.1.0 through 11.1.3 FP1 can return detailed technical error messages to the browser, allowing remote, unauthenticated attackers to harvest implementation details about the affected system (CWE-209). Per the CVSS 3.1 vector AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N, exploitation is network-reachable and requires no credentials or user interaction, but the impact is confined to low-value information disclosure that is most useful as reconnaissance for follow-on attacks rather than for direct compromise. No public exploit code has been identified and the issue is not listed in CISA KEV, so this is a genuine but low-priority weakness whose practical risk largely depends on whether the Controller web interface is reachable from untrusted networks.
Denial of service in OpenImageIO prior to 3.1.16.0: processing a crafted ZIP/Deflate-compressed TIFF while TIFF multithreading is enabled can crash the image-reading process. The bug is an object destruction-order error in TIFFInput::read_native_scanlines(), where an error path returns before queued asynchronous strip-decompression tasks are joined, letting worker tasks touch stale stack and heap storage (use-after-scope). The assessed vector is CVSS:3.1/AV:L/AC:H/PR:N/UI:R/S:U/C:N/I:N/A:H - no privileges required, but a user or pipeline must open the malicious file (UI:R) and the crash depends on a narrow timing window (AC:H); impact is availability only, with no memory disclosure or code execution, and no public exploit identified at time of analysis.
Server-side request forgery in SveltyCMS 0.0.6 lets an authenticated, low-privileged user (CVSS PR:L, confirmed in both the supplied CVSS 4.0 vector and the independent assessment) force the server to make outbound requests by supplying a crafted remote media/importer URL to the File Upload Endpoint at /mediagallery/upload-media. Exploitation requires valid credentials on a reachable instance that exercises remote media fetching, and the resulting impact is limited (low confidentiality, integrity, and availability, no scope change, no user interaction) - this is a real-but-modest issue rather than a top-priority one. No public exploit code was identified at time of analysis, the flaw is not confirmed actively exploited (CISA KEV), and the vendor has published an upstream fix commit (05b4f9efeb79e9d72a693232334d7529687f896f) that replaces raw outbound fetches with validated egress helpers.
Improper access control in SveltyCMS 0.0.6 allows an authenticated low-privilege user to submit privileged attributes such as role, isAdmin, roleIds, or permissions through the User Attribute Update Endpoint at src/routes/api/[...path]/+server.ts and thereby alter account privileges. The flaw is remotely reachable with no user interaction and, per the independent assessment, has high confidentiality, integrity, and availability impact, though exploitation requires valid low-privilege credentials and the product is a pre-release CMS with essentially no production deployment. No public exploit code was identified at time of analysis and CISA KEV does not list this CVE; an upstream fix is available as commit 05b4f9efeb79e9d72a693232334d7529687f896f, but a tagged release containing it is not independently confirmed.
IBM WebSphere Application Server 9.0 and 8.5 allow remote, unauthenticated attackers to obtain sensitive file-system information through the FileTransfer servlet due to incorrect authorization (CWE-863). Exploitation requires network access to the servlet endpoint, which is part of the administrative/deployment interface, and impact is limited to confidentiality (CVSS 5.3, C:L). No public exploit code or active exploitation has been identified at time of analysis, but a vendor patch is available per IBM's advisory.
Denial of service in Netty's HTTP/2 HPACK encoder (HpackEncoder) allows an unauthenticated remote attacker to force excessive CPU and memory consumption by sending SETTINGS frames that advertise an oversized MAX_HEADER_TABLE_SIZE, causing the encoder to retain an excessive number of unique header entries. The issue affects applications and Red Hat products that embed Netty and expose an HTTP/2 listener - including JBoss EAP 7 and 8, Keycloak, Red Hat Single Sign-On 7, AMQ Broker 7, Data Grid 8, Fuse 7, Quarkus, Camel 4 for Quarkus 3, Camel for Spring Boot 4, Apicurio Registry 3, and Debezium 3 - while HTTP/1.1-only endpoints are not exposed. Impact is bounded to availability (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L, CVSS 5.3) with no data disclosure or modification; no public exploit code has been identified at time of analysis and there is no confirmed active exploitation.
Unauthenticated network access to the SOAP/JMX administrative connector on IBM WebSphere Application Server 8.5 and 9.0 can bypass authentication, allowing an attacker to read limited information from the server (CWE-306, CVSS 5.3). The flaw is remotely reachable without credentials or user interaction (PR:N/UI:N/AC:L), but impact is confined to low confidentiality loss - no code execution, configuration change, or denial of service. A vendor patch is available per IBM's advisory; no public exploit has been identified at time of analysis, and real-world exposure is constrained by the fact that the SOAP/JMX connector is normally bound to a management interface that hardened deployments segment or firewall from untrusted networks.
Time-based blind SQL injection in CordysCRM versions prior to 1.7.0 allows an authenticated attacker holding the MODULE_SETTING:UPDATE permission to inject SQL into the ORDER BY clause via the sort.name field of the POST /account-pool/page endpoint, enabling inference of database metadata and sensitive values through induced delays and causing service degradation. Exploitation requires a privileged, trusted account and the blind nature of the flaw makes data extraction slow, so this is a genuine but modest-priority issue rather than a mass-exploitable one. No public exploit code has been identified at time of analysis, and the flaw is fixed in release 1.7.0 (GHSA-xrcr-hj37-q83j).
Mealie versions before 3.21.0 expose an insecure direct object reference (IDOR) in the per-user ratings and favorites endpoints, allowing any authenticated account holder to read another user's recipe ratings, favorite flags, and the private recipe identifiers attached to them by simply substituting a victim's user UUID in the request URL. Exploitation requires a valid account on the target instance but does not require the attacker to share a group or household with the victim - cross-group and cross-household reads succeed, which the upstream regression tests explicitly verify by expecting 403 once the fix is applied. The impact is read-only information disclosure (CVSS 3.1 5.3, AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:N/A:N), with no write capability or service disruption, and no public exploit code identified at time of analysis; the practical limiting factor is that the attacker must know or enumerate target user UUIDs.
An out-of-order FTP command sequence can blind Suricata's FTP inspection for the remainder of a TCP flow. In Suricata versions prior to 8.0.6, when an FTP client issues a RETR or STOR command before completing PORT or PASV negotiation, the parser in src/app-layer-ftp.c raises an unrecoverable application-layer error rather than flagging a recoverable protocol event; that fatal state disables FTP parsing for the rest of the flow, creating a reporting and rule-matching blind spot in IDS deployments while IPS deployments drop the affected flow outright. Remote attackers who can influence the FTP control channel visible to a sensor (CVSS 3.1 base 3.7, AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:L/A:N, CWE-755) can therefore evade parser-dependent rules and logging, but no target host is compromised and no confidentiality is lost; no public exploit code or confirmed active exploitation was identified at time of analysis, and the issue is fixed in 8.0.6.
SIP body-length truncation in Suricata 8.0.0 through 8.0.5 causes the IDS/IPS engine to expose only the first 65,536 bytes of oversized SIP request or response bodies to frame-based inspection, letting an unauthenticated remote attacker hide malicious content in the omitted remainder and evade detection. The flaw resides in the Rust SIP parser, which stores body offsets and lengths in 16-bit fields, so any body larger than 65,536 bytes wraps the stored length; this only matters on deployments that enable SIP application-layer parsing and run rules referencing frame:request.body or frame:response.body, and it degrades detection integrity only (no crash, no code execution, no data exposure). Per the CVSS vector (AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:L/A:N, score 3.7) the impact is limited to integrity of the detection pipeline, and no public exploit code or confirmed active exploitation was identified at time of analysis.
SiYuan versions through 3.8.4 fail to enforce authorization on the getDynamicIcon endpoint, allowing authenticated attackers holding a read-only publish token to disclose metadata-such as block titles, names, aliases, and hierarchical paths-of restricted documents via a template injection triggered by type=8 requests. The flaw is rated medium severity (vendor CVSS 5.3, independent CVSS 3.1 5.3) and impacts confidentiality only; no public exploit or active exploitation has been identified at time of analysis.
Open redirect in Cotonti through 1.0.0 lets remote, unauthenticated attackers bounce victims to arbitrary attacker-controlled sites by supplying a hostname that merely begins with the legitimate site domain, defeating the cot_url_check() validation used by the ratings plugin and other redirect callers. Exploitation is constrained by required victim interaction - a user must follow a crafted link through a reachable redirect caller such as plugins/ratings/ratings.ajax.php - and yields only redirection to an attacker host, not code execution or data compromise, consistent with the assessed CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:N/I:L/A:N vector (CVSS 4.0 base 5.3). This is a low-severity, internally consistent issue with no public exploit identified at time of analysis; there is no evidence of active exploitation or KEV listing.
Log injection in IBM WebSphere Application Server 9.0 and 8.5 lets remote, unauthenticated attackers corrupt server log content by supplying a crafted LTPA token cookie that the application writes to its logs without neutralizing control characters. The impact is limited to integrity (CVSS 3.1 base score 3.7, I:L) with no confidentiality or availability loss and no code execution, and exploitation is rated high complexity because the injected content must survive into the log in a form a human or downstream log consumer then misinterprets. No public exploit code and no confirmed active exploitation (CISA KEV) were identified at time of analysis; a vendor patch is available per the IBM advisory.
HCL BigFix Service Management exposes internal IP addresses through its own HTTP responses - response headers, error messages, or rendered page content - allowing anyone able to reach the instance to harvest RFC1918 addressing and infer the internal network topology behind it. The consequence is reconnaissance only: no data modification, no service disruption, and a low vendor base score of 3.1, making this a low-priority item rather than an urgent one. No public exploit code was identified at time of analysis and there is no KEV or EPSS signal in the supplied data. The vendor vector specifies PR:L (a valid session) while the description claims an unauthenticated attacker, so authentication requirements are not confirmed from available data; the vendor's AC:H also implies an additional non-trivial condition or timing is needed to elicit the leak.
Cross-site request forgery in the Cotonti CMS contact plugin (through version 1.0.0) lets an attacker auto-submit the contact form from an attacker-controlled page, sending forged messages into the administrator inbox that appear to originate from a logged-in victim (CWE-352, CVSS 4.0 base 5.3 / assessed CVSS 3.1 5.3). The submission handler never validates an anti-CSRF token, so any authenticated user who visits a malicious page while their session is active can be silently impersonated. Exploitation requires victim interaction and an enabled contact plugin, and the impact is limited to message forgery and inbox spam - no data disclosure or system compromise; no public exploit code and no confirmed active exploitation were identified at time of analysis.
Cotonti through 1.0.0 allows unauthenticated remote attackers to forge rating submissions on behalf of an authenticated user when the ratings plugin is enabled. The attacker must lure a logged-in user with rating privileges to a malicious page that auto-submits a POST to the ratings AJAX handler, modifying stored rating data without the victim’s intent. A vendor patch is indicated and an upstream fix exists in GitHub PR 1900, but no public exploit was identified at time of analysis.
Missing integrity verification in spatie Scotty's self-update handler (SelfUpdater::update in app/Updater/SelfUpdater.php) allows a tampered release artifact to be downloaded and installed as the running application in versions up to and including 1.4.2. Because the updater fetches the phar without validating a signature or hash, an attacker who can substitute the distributed artifact - for example through a compromised GitHub release or maintainer credential, a malicious mirror, or interception of the download - gains execution of arbitrary PHP within the Scotty process once an operator triggers the self-update; CVSS 4.0 lists PR:N (unauthenticated) with UI:P, so user interaction is required but no prior authentication is. No public exploit code has been identified at time of analysis and the issue is not listed in CISA KEV; the vendor-supplied CVSS 4.0 base score is 5.3, while our independent assessment rates it higher under CVSS 3.1 (AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H) because of the high integrity and confidentiality impact of replacing the application binary.
Server-side request forgery in ArcadeDB versions before 26.9.1 allows an authenticated user to bypass the built-in SSRF guard by supplying hostnames that resolve to IPv6 transition addresses (NAT64, 6to4, or Teredo) which embed RFC 1918 or loopback IPv4 targets, causing IMPORT DATABASE and server commands to fetch internal-only resources on the attacker's behalf. Exploitation requires valid credentials and permission to invoke the URL-fetching command, and meaningful impact depends on reachable internal services - most notably a cloud instance metadata endpoint such as 169.254.169.254 - being network-accessible from the ArcadeDB host; the assessed vector (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:L/I:L/A:N) reflects low confidentiality and integrity impact with a scope change and no availability impact. No public exploit identified at time of analysis, and there is no confirmed active exploitation (CISA KEV); the vendor advisory notes that patches are available, making this a medium-priority credentialed issue rather than an urgent one.
An authenticated, low-privilege ArcadeDB user can bypass per-type record ACLs and durably append graph edges to protected vertex types in versions before 26.9.1. The flaw is an incomplete fix of GHSA-c23x-pqcj-7hfm: the authenticated principal is bound only on the HTTP handler thread, while the async worker threads used by the parallel edge-connect phase of POST /api/v1/batch/{database} carry no current user, so LocalDatabase.checkPermissionsOnFile returns early and the CREATE_RECORD/UPDATE_RECORD checks on the target vertex type are never enforced. The attacker must already hold CREATE_RECORD on some edge type and submit an edge-load batch with parallelFlush left at its default of true; setting parallelFlush=false causes the request to be correctly rejected. Impact is confined to integrity of graph adjacency - no protected data is read and there is no denial of service - and no public exploit identified at time of analysis; a vendor patch is available in 26.9.1.
Unauthenticated remote attackers can query the Desktop API in Synology DiskStation Manager (DSM) to retrieve non-sensitive information, affecting DSM versions before 7.2.1-69057-10, 7.2.2-72806-7, and 7.3.2-86009-2. The CVSS 3.1 score is 5.3 (AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N), and while no public exploit code or active exploitation has been identified at time of analysis, the flaw still offers reconnaissance value because no authentication or user interaction is required. Given the limited confidentiality impact and lack of integrity or availability effects, this is a low-priority issue, with risk primarily determined by whether the DSM management interface is internet-facing.
Improper encoding and escaping of output in the Auth API of Synology DiskStation Manager (DSM) lets remote, unauthenticated attackers retrieve information the vendor classifies as non-sensitive, affecting deployments before 7.2.1-69057-12, 7.2.2-72806-9, 7.3.2-86009-4, and 7.4-90075. The CVSS vector (AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N, 5.3) confirms exploitation needs no credentials, no user interaction, and no non-default configuration, so the only meaningful barrier is impact rather than access. There is no public exploit identified at time of analysis and no KEV listing, and because disclosure is limited to non-sensitive information, the practical value to an attacker is reconnaissance and fingerprinting rather than direct compromise; vendor-released patches exist in the fixed DSM builds published under Synology advisory SA_26_13.
Plaintext third-party AI provider API keys can be extracted from any WordPress site running the GPTranslate - Multilingual AI Translation Agent plugin at version 2.34.6 or earlier, because the plugin emits the stored credential into a public front-end JavaScript asset loaded by enqueue_frontend_scripts. The exposure is unauthenticated and requires no user interaction - an attacker only needs to fetch any public page and apply the inverse transformation that is itself bundled in the plugin's own public JavaScript to recover the key in cleartext (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N, 5.3 Medium as published; our assessment raises the confidentiality component to High given that a raw billing-capable credential is disclosed, yielding 7.5). Keys for OpenAI, DeepL, xAI/Grok, Gemini, Claude, and Google Cloud Translation are all affected, and the default configuration (gpt-3.5-turbo in client mode) is vulnerable; only deepseek-* models and gpt-* models running in server-proxy mode correctly suppress key emission. No public exploit code was identified at time of analysis and the vulnerability is not listed in CISA KEV, though the recovery technique is trivially derivable from the plugin's own shipped assets, so treat exploitation as plausible and rotate credentials.
Unauthenticated arbitrary shortcode execution in the Divi theme for WordPress (all versions up to and including 5.11.1) allows remote attackers to invoke any registered shortcode by supplying an unvalidated value to do_shortcode through an action endpoint that lacks a proper authorization check (CWE-862). Rated CVSS 5.3 (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:N), the confirmed impact is integrity-only: no confidentiality or availability loss and no code execution is claimed. When the non-default 'Force Enable D4 Shortcode Framework' option is enabled, attackers can additionally abuse the et_pb_contact_form shortcode to send email to an attacker-chosen recipient with attacker-controlled content; sites without that option still permit arbitrary shortcode execution but not this specific email vector. No public exploit code was identified at time of analysis and the issue is not listed in CISA KEV, so this is a genuine but moderate-severity weakness rather than an actively exploited one.
Stored cross-site scripting in Aureus ERP through 1.6.0 lets an authenticated user who holds the payment-term create permission (create_account_payment::term) persist arbitrary JavaScript in the Payment Term note field of the Accounting plugin, which then executes in the browser of any other user who opens that Payment Term record. Exploitation requires a privileged account plus a victim opening the affected infolist view (CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:L/I:L/A:N, assessed base 5.1 under CVSS 4.0), so impact is bounded by that privilege gate and the required user interaction; there is no public exploit identified at time of analysis and no CISA KEV listing. An upstream fix exists as GitHub pull request #1562, which sanitizes the note field, but no tagged patched release is confirmed in the provided data.
Cotonti CMS (often branded simply as Cotonti) through version 1.0.0 stores page bodies beginning with the `redir:` prefix as redirects without validating the destination, letting an authenticated user who can create or edit pages point visitors from a trusted Cotonti-hosted page to an arbitrary external site for phishing. Exploitation requires a valid account holding page creation or edit permissions (CVSS vector PR:L) and depends on a victim opening the crafted page (UI:P), with impact limited to redirecting visitors - no server-side code execution or data compromise is possible. There is no indication of active exploitation in the wild and no public exploit code identified at time of analysis; EPSS and KEV data were not provided. The vendor/upstream fix is PR #1901, which adds a `cot_url_check` guard rejecting external `http`/`ftp` destinations.
Stored cross-site scripting in SveltyCMS 0.0.6 allows an authenticated user who can upload media to plant a malicious SVG that executes in the browser of anyone who later views it. The flaw sits in src/utils/media/media-service.server.ts, where the large-file/streaming upload path bypassed the sanitizeSvg / bufferAndSanitizeSvg routine and the MAX_SVG_BYTES size cap, so unsanitized SVG markup reaches storage and is served back to victims; the supplied vector (AV:N/PR:L/UI:P) confirms exploitation requires a media-upload-capable account and victim interaction, and no public exploit code or confirmed active exploitation was identified at time of analysis. Scored 5.1 in CVSS 4.0 terms, this is a moderate-severity stored XSS confined to SveltyCMS 0.0.6 prior to the referenced fix commit.
Stored cross-site scripting in the profile management functionality of T-Systems' TAO 2.0 suite lets an authenticated user plant malicious HTML or JavaScript inside their own personal-data fields, which is then persisted unsanitised and rendered to whoever views that profile - including administrative staff. Because the payload executes in the viewer's browser under their own session context, the attacker can read session-available information and perform actions with the victim's permissions; server-side data remains out of scope per the assessed vector (S:C with C:L/I:L only on the subsequent system). Exploitation requires a valid TAO 2.0 account with permission to edit profile fields (PR:L) plus a second party actually opening the tainted profile (UI:P), so it is a real but modest-priority issue rather than a self-propagating or unauthenticated attack, and no public exploit code or active exploitation is identified at time of analysis.
Unauthenticated tenants can read other organizations' uploaded files in CordysCRM (1Panel-dev) versions prior to 1.7.4, because the Shiro-configured anonymous routes /attachment/preview/{id} and /pic/preview/{id} resolve attachments through AttachmentService.getResource with a bare primary-key lookup and no ownership, organization, or permission check. Exploitation requires the attacker to know, guess, or observe a valid IDGenerator.nextStr file id; the ids are not sequentially predictable, which raises attack complexity, but ids leaked through shared preview links, logs, or browser history permit direct retrieval. The issue is read-only disclosure of stored files and is fixed in 1.7.4; no public exploit code or confirmed in-the-wild exploitation was identified at time of analysis. This is a genuine unauthenticated broken-access-control (IDOR) flaw but a moderate real-world priority, not a critical emergency.
Cross-tenant data exposure in HCL BigFix Service Management allows an authenticated user with a valid account in one tenant to read restricted data elements belonging to other tenants because of improper access controls. Affected deployments are those operated in shared or multi-tenant mode; the impact is limited to confidentiality (unauthorized viewing), with no modification or disruption of data, and no user interaction is required on the victim's side (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:L/I:N/A:N, base 5.0). Exploitation requires a legitimate authenticated foothold in at least one tenant, so this is not remotely exploitable by unauthenticated attackers, and no public exploit code or confirmed active exploitation (CISA KEV) was identified at time of analysis.
Local information disclosure in cockpit-machines on Red Hat Enterprise Linux 9 and 10 allows an authenticated local attacker who can enumerate other users' process arguments to read guest VM installation credentials such as rootPassword and userPassword. The credentials are passed by install_machine.py as a JSON command-line argument during VM creation, so exposure is limited to the window while an administrator's installation workflow is running and depends on the host's process-visibility policy (e.g. /proc without hidepid=2). Publicly available exploit code exists, but there is no confirmed active exploitation; the issue requires local access plus a concurrently active install and cannot be triggered remotely or unauthenticated.
Local information disclosure in the cockpit-machines package on Red Hat Enterprise Linux 9 and Red Hat Enterprise Linux 10 lets an unprivileged local user capture a Red Hat Subscription Management (RHSM) offline token that is passed as a command-line argument to a helper script during token validation. Rated 5.0 (CVSS:3.1/AV:L/AC:L/PR:L/UI:R/S:U/C:H/I:N/A:N), the impact is confidentiality-only: a local account that can read other processes' metadata (for example /proc/<pid>/cmdline via ps) must catch the brief window in which a privileged operator runs the validation, and the stolen token can then be exchanged for RHSM access tokens. There is no public exploit code identified at time of analysis and no CISA KEV entry; Red Hat tracks the issue as CVE-2026-92745 (EUVD-2026-83260).
Authenticated SQL injection in the NEX-Forms - Ultimate Forms Plugin for WordPress (all versions up to and including 9.3.0) allows a user holding the plugin's 'custom-level' capability or above to inject arbitrary SQL through the 'additional_params' parameter and read sensitive data out of the underlying WordPress database. The injection sits in the plugin's records/dashboard query path, where the tainted array is forwarded from get_table_records() into get_total_records() without the operator allowlist being re-applied to the SQL sink, and no user interaction or non-default configuration is required beyond the plugin being installed. Impact is confidentiality-only - an attacker can extract data such as user records and other stored content, but cannot modify or destroy it, and no public exploit code and no confirmed active exploitation were identified at time of analysis; this is a genuine but mid-tier priority issue that is primarily constrained by the high privilege level required.
Arbitrary file read in the WordPress plugin Store Exporter (woocommerce-exporter) through version 2.8.0 allows an authenticated attacker holding shop-manager-level access or above to supply a crafted 'filename' parameter and retrieve the contents of any file readable by the web server. The traversal is confined to disclosure - there is no write, no code execution, and no path to privilege escalation beyond the account already required - and the high-privilege prerequisite (CVSS PR:H, score 4.9) keeps this a genuine but low-priority issue rather than an urgent one. No public exploit identified at time of analysis, and the issue is not listed in CISA KEV; exploitation is only reachable from the plugin's export handler by a user who already holds elevated store-management rights.
HTTP request smuggling in IBM WebSphere Application Server and WebSphere Application Server Liberty - as bundled in IBM CICS TX Advanced per the affected CPE - lets remote attackers desynchronize request framing between a front-end intermediary and the WebSphere back end, enabling request tampering or response poisoning. Exploitation requires the server to sit behind a reverse proxy, load balancer, or CDN that parses Content-Length/Transfer-Encoding framing differently than the back end; no authentication or user interaction is needed (PR:N/UI:N), but attack complexity is high (AC:H) because the attacker must find and reliably trigger a specific parser discrepancy in the real deployed chain, and a directly exposed server with no divergent front-end parser is not smuggle-able. Impact is bounded to low confidentiality and integrity with no availability effect (CVSS 3.1 base 4.8); this is a genuine but modest-severity issue rather than a top priority, and no public exploit was identified at time of analysis.
HTTP request smuggling in IBM WebSphere Application Server Liberty - the runtime bundled with IBM CICS TX Advanced 10.1 - lets remote, unauthenticated attackers desynchronize HTTP message framing between an upstream HTTP parser and the Liberty backend, so a smuggled request can be processed as if it came from the trusted front-end, with low confidentiality and integrity impact on subsequent requests sharing the connection. Exploitation requires a multi-tier topology (reverse proxy, load balancer, or CDN in front of or behind Liberty) where the two parsers disagree on Content-Length/Transfer-Encoding or chunked framing, and is rated high complexity because the specific parsing gap must be discovered and precisely crafted while connection reuse must be achievable; no user interaction is required. Risk is moderate and consistent with the vendor CVSS of 4.8 (Medium); no public exploit identified at time of analysis, and patch is available from the vendor.
HTTP request smuggling (CWE-444) in IBM WebSphere Application Server 8.5, 9.0, and Liberty lets an authenticated, low-privileged remote attacker desynchronize HTTP request framing between WebSphere and an intermediary so that one user's request boundaries are misattributed on a shared connection. The issue only manifests when WebSphere sits behind or in front of a second HTTP parser (reverse proxy, load balancer, or CDN) that disagrees with WebSphere on Content-Length versus Transfer-Encoding chunked handling; a standalone WebSphere instance with no fronting or backend HTTP hop is not affected, and the 4.2 score reflects high attack complexity, partial confidentiality and integrity impact, and no availability or scope change. No public exploit code and no confirmed active exploitation were identified at time of analysis, and a vendor patch is available per IBM's advisory.