Command injection in rpm's `rpmbuild -t*` subcommand allows arbitrary command execution with build-user privileges when an attacker controls the path or filename of a tarball being processed. The vulnerability arises because shell metacharacters in tarball names are not sanitized before being passed to a shell invocation inside rpmbuild. Exploitation is most realistic in automated CI/CD pipelines that fetch externally supplied artifact archives and pass them directly to rpmbuild, a common pattern in open-source package maintainer workflows. No public exploit code or active exploitation (CISA KEV) has been identified at time of analysis.
Command injection in rpmuncompress allows a local attacker to execute arbitrary commands under the invoking user's context by supplying an archive file whose filename contains shell metacharacters. Affecting all actively supported Red Hat Enterprise Linux versions (6 through 10) and Red Hat Hardened Images, the flaw is triggered when a user or automated build pipeline invokes rpmuncompress on a maliciously named archive. No public exploit or active exploitation has been identified at time of analysis, but the low attack complexity and broad RHEL enterprise deployment footprint make this a meaningful risk for organizations processing untrusted archives in development or CI/CD workflows.
Privilege escalation in util-linux's nsenter --join-cgroup option allows an unprivileged local user to move and terminate root-owned host processes by exploiting a persistent file descriptor opened as root that survives across namespace and credential changes. The kernel evaluates cgroup migration requests against the credentials held at fd-open time rather than the current caller's credentials, so any program that inherits the still-open cgroup.procs fd effectively retains root's cgroup management authority. Red Hat Enterprise Linux 7-10, Red Hat Hardened Images, and Red Hat OpenShift Container Platform 4 are confirmed affected; no public exploit has been identified at time of analysis.
Broken object-level authorization in the Timetics WordPress plugin through version 1.0.61 allows authenticated users holding the plugin's custom staff role to modify, disable, or take over appointments belonging to other staff members via the REST API. The endpoint accepts appointment identifiers without validating ownership, enabling lateral privilege abuse among staff peers. A public WPScan PoC confirms exploitability, though real-world impact is bounded by the PR:H prerequisite of a pre-existing staff account.
Cross-instructor content tampering in MasterStudy LMS WordPress Plugin before 3.7.46 allows authenticated instructor-role users to delete or modify curriculum sections and materials belonging to courses owned by other instructors. The root cause is CWE-639 (Authorization Bypass Through User-Controlled Key): the plugin performs operations on curriculum objects without confirming the requesting instructor owns the parent course. A publicly available proof-of-concept exists via WPScan; no active exploitation (CISA KEV) has been recorded, and the CVSS score of 3.8 reflects the restricted impact scope.
Local privilege escalation in Elastic Agent on Windows enables a low-privileged local user to achieve SYSTEM-level code execution by exploiting overly permissive access controls (CWE-732) on resources created during unprivileged-mode installation. The agent service, running as SYSTEM, will execute attacker-controlled binaries placed via the CAPEC-642 binary replacement technique. Vendor-released patches 8.19.21, 9.4.6, and 9.5.2 are available per advisory ESA-2026-150; no public exploit identified at time of analysis.
Local privilege escalation in util-linux's SUID mount(8) binary exposes all major Red Hat Enterprise Linux versions (7-10), Red Hat Hardened Images, and OpenShift Container Platform 4 to root-level file ownership and permission manipulation by low-privileged local users. The flaw stems from a time-of-check to time-of-use (TOCTOU) race: mount(8) reads the bind-mount source from fstab but does not pin the resolved inode before issuing the privileged syscall, allowing an attacker with write access to the source path or a writable ancestor to substitute an arbitrary host directory. When fstab entries additionally specify X-mount.owner, X-mount.group, or X-mount.mode, root subsequently applies chown or chmod to the attacker-redirected path, completing the privilege escalation. No public exploit code or CISA KEV listing is identified at time of analysis.
OS command injection in Westermo WeOS 5.x (versions 5.24 and later) allows an authenticated administrator to execute arbitrary shell commands on the underlying device operating system by supplying a crafted media definition through the management interface. The scope change component of the CVSS vector confirms exploitation breaks out of the management application context into the OS layer, enabling full device compromise. No public exploit code or CISA KEV listing has been identified at time of analysis; however, CISA issued ICS advisory ICSA-25-261-01, elevating concern for operators in critical infrastructure sectors.
Unauthenticated schema disclosure in the Rank Math SEO WordPress plugin (versions 1.0.48 through 1.0.276) allows any anonymous visitor to read the structured schema markup and associated post content of posts the site owner has restricted - including drafts, pending review, private, scheduled, and password-protected posts. The plugin renders JSON-LD schema on the front end without verifying whether the requesting user is authorized to view the underlying post, effectively leaking embargoed or confidential content. A publicly available exploit is documented by WPScan; vendor patch version 1.0.277 is available. No active exploitation confirmed in CISA KEV at time of analysis.
Template injection in Eclipse Ditto's ImplicitThingCreationMessageMapper (versions 1.3.0 through 3.9.6) allows authenticated connection publishers to override the administrator-configured access policy of a newly provisioned digital twin by embedding JSON-breaking characters in attacker-controlled inbound message headers. The mapper substitutes header placeholder values as raw, unescaped strings into a pre-configured JSON thing-template before parsing, enabling a crafted header value containing a double-quote to escape its string context and inject an inline _policy object - effectively granting the attacker full read/write access to the twin while permanently revoking the legitimate owner's access. Exploitation requires a non-default mapper configuration and a template that reflects attacker-controllable headers; no public exploit code or active exploitation (CISA KEV) has been identified.
OS command injection in Jenkins TICS Plugin 2025.1.1 and earlier allows an authenticated, low-privileged Jenkins user who can control build environment variable values to execute arbitrary commands on the build agent hosting the job. The flaw stems from unsanitized environment variable values being passed directly to OS command invocations within the plugin, and the scope change (S:C) means impact falls on the agent node rather than the Jenkins controller. No public exploit code exists and SSVC rates exploitation status as none with non-automatable attack path, placing real-world priority lower than the raw CVSS score suggests.
Missing authorization on the `set_course_mode_price()` instructor endpoint in Open edX Platform allows any authenticated user - including learner accounts with zero course roles - to overwrite the honor-mode price and currency for any course on the platform. The view at `lms/djangoapps/instructor/views/instructor_dashboard.py:430` enforces only `@login_required` with no course-level role check, and became an orphaned API route after its companion frontend modal was removed. No public exploit code has been identified at time of analysis, but the low barrier (any valid session) makes automated mass-price manipulation trivially scripted against internet-facing LMS instances.
Unauthenticated remote denial of service in Libreswan's pluto IKE daemon allows any network attacker to repeatedly crash and restart the VPN service by sending a malformed X.509 certificate in a CERT payload during IKE handshake negotiation. The crash is triggered by a reachable assertion in `add_decoded_cert()`: when NSS's `CERT_ExtractPublicKey()` receives a certificate with an invalid RSA exponent (e.g., zero), it returns NULL, which the FIPS-mode code path does not handle, causing the daemon to abort. The vendor explicitly confirms no remote code execution is possible; impact is limited to availability. No CISA KEV listing and no public exploit code have been identified at time of analysis.
Path traversal in the RadImageEditor component of Progress Telerik UI for ASP.NET AJAX (all versions prior to v2026.3.812) allows unauthenticated remote attackers to read arbitrary files from the server by supplying crafted state values that manipulate the control's image cache lookup. The vulnerability stems from insufficient server-side validation of client-controlled input that determines which file the cache mechanism retrieves, enabling directory traversal beyond the intended image directories. No public exploit has been identified at time of analysis, but the low-complexity unauthenticated attack vector makes this a practical risk for any internet-exposed ASP.NET application using RadImageEditor.
Memory exhaustion via unbounded decompression in HTTPX2 (all versions before 2.12.0) allows any HTTP server the client contacts to cause severe memory pressure or out-of-memory process termination by returning a crafted compressed response body. The content decoders in `_decoders.py` call `decompress()` without a size cap, so a single 64 KiB compressed chunk can inflate to approximately 64 MiB in one allocation across gzip, deflate, brotli, and zstd encodings - all enabled by default. No active exploitation is confirmed in CISA KEV, and no public proof-of-concept is referenced, but the decompression-bomb technique is well understood and trivially reproducible.
Script execution enforcement failure in Sonatype Nexus Repository 3 allows privileged authenticated users to invoke pre-existing Groovy scripts via POST /service/rest/v1/script/{name}/run even after administrators explicitly set nexus.scripts.allowCreation=false to disable the scripting subsystem. The execution endpoint did not consult the administrative flag, meaning a hardening action administrators believed fully blocked script activity was silently ineffective against execution of pre-existing scripts. A vendor-released patch is available in Nexus Repository 3.96.0; no active exploitation has been confirmed in CISA KEV and no public proof-of-concept has been identified.
URI authority injection in fast-uri allows an attacker-supplied port value containing delimiter characters to demote the intended host to userinfo and redirect the parsed authority to an attacker-controlled hostname, bypassing downstream re-validation because both fast-uri and Node.js's built-in URL parser silently accept the malformed output as well-formed. Applications in the Fastify ecosystem that build URIs from parts — passing untrusted data into the port field of a URI object through serialize, normalize, or equal — on fast-uri versions before 2.4.6, 3.1.7, or 4.1.4 are affected. No public exploit has been identified at time of analysis, and the CVE is not listed in CISA KEV; vendor-released patches are available across all three active release lines.
Host-parsing ambiguity in fast-uri allows attackers to silently bypass SSRF denylists, redirect allowlists, and proxy routing rules by submitting URLs containing unbalanced authority brackets. Affected versions 2.4.5, 3.1.6, and 4.1.3 of the Fastify ecosystem URI parsing library return the malformed bracket-containing string as the host without raising an error, while Node.js's URL implementation and downstream HTTP clients resolve the same string to an entirely different host. Any application that reads the fast-uri-parsed host to enforce a security policy, then passes the original URL to an HTTP client, evaluates its controls against a string that is never the host the request actually reaches.
Heap-use-after-free in libcurl versions 8.14.0 through 8.21.0 allows remote unauthenticated attackers to crash applications performing concurrent TLS transfers via the multi interface. When libcurl builds against OpenSSL 3 with provider configurations, an allocated OpenSSL library context is tied to an easy handle's lifetime without an ownership reference being passed to OpenSSL; if the easy handle is destroyed while a pooled TLS connection remains active, subsequent I/O or post-handshake operations dereference the freed context, triggering a heap-use-after-free. No public exploit or active exploitation (CISA KEV) has been identified at time of analysis, and EPSS remains low at 0.22% (12th percentile).
Secure-flag stripping in curl 8.13.0 through 8.21.0 allows a server to cause curl to transmit session cookies over unencrypted HTTP by returning a Set-Cookie header that places a horizontal tab character (ASCII 9) instead of a space immediately before the Secure attribute. curl's cookie parser fails to recognise the attribute under this condition, stores the cookie without its Secure constraint, and subsequently includes it in plaintext HTTP requests to the same host. No public exploit code has been identified and EPSS probability is 0.22% (12th percentile), reflecting low observed exploitation activity despite the straightforward triggering mechanism.
Connection pool mismanagement in libcurl silently reuses HTTPS connections across requests that differ in their `CURLSSLOPT_NATIVE_CA` (Native CA Store) SSL option, affecting all curl releases from 7.71.0 through 8.21.0. When an application uses a shared connection pool to the same hostname with mixed native CA settings, libcurl fails to distinguish connections by their CA trust configuration, potentially applying the wrong certificate authority validation to a subsequent request. No public exploit code exists and EPSS sits at 0.21% (12th percentile), indicating low active exploitation probability, though the vulnerability's broad version footprint - spanning several years of releases - makes it relevant wherever libcurl is deployed with mixed native CA configurations.
Stored Cross-Site Scripting in Icegram Express WordPress plugin versions before 5.8.6 allows administrator-level users to inject malicious scripts into HTML attributes via the list description setting. When another user subsequently loads the affected page, the injected payload executes in their browser, enabling session hijacking, credential capture, or unauthorized actions on their behalf. A publicly available proof-of-concept exploit has been confirmed by WPScan, though real-world impact is substantially constrained by the high-privilege prerequisite required to plant the payload.
Connection deadlock in golang.org/x/crypto/ssh (all versions before 0.56.0) allows a malicious remote SSH peer to permanently freeze an SSH connection by flooding the incomingRequests queue of a channel that has been registered in the mux's chanList but not yet established. The attack exploits the absence of any established-state guard in the packet handler, causing a deadlock that renders the entire connection unresponsive without triggering an explicit teardown. No public exploit has been identified and no CISA KEV listing exists, but the network-accessible, no-authentication CVSS vector makes this straightforward to trigger against any Go application using this library for SSH.
Denial-of-service in the Go extended cryptography library's SSH package allows a malicious peer to deadlock an entire SSH connection by sending crafted RFC 4254 channel messages after a channel is established. All versions of golang.org/x/crypto/ssh below 0.56.0 are affected; the flaw manifests because unrecognized channel messages were buffered and allowed to block rather than being rejected as protocol errors. No public exploit or active exploitation has been identified, but the attack requires no privileges in the worst-case client scenario where a malicious SSH server targets a connecting client.
CSS injection in GutenKit WordPress plugin (versions before 2.5.1) allows any authenticated user holding the Contributor role or above to embed arbitrary stylesheets into posts served to all site visitors and anonymous users. The plugin fails to validate or escape style settings before interpolating them into CSS output on the front end, enabling defacement, interface redressing, and forced loading of external resources. JavaScript execution is explicitly ruled out at the Contributor privilege tier, capping the attack surface below full cross-site scripting. A publicly available exploit has been disclosed by WPScan.
Public key pinning enforcement in libcurl (versions 7.45.0 through at least 8.21.0) is silently bypassed when CURLOPT_PINNEDPUBLICKEY is combined with CURLOPT_SSL_VERIFYPEER=0 and CURLOPT_SSL_VERIFYHOST=0 - the exact configuration adopted by applications that rely on pinning as their sole TLS trust anchor after disabling standard certificate chain validation. When a server presents no certificate at all, libcurl's pin-check code path is never reached and the connection succeeds, enabling a network-positioned attacker to conduct MITM attacks against applications that believed pinning would protect them. No public exploit is identified at time of analysis and EPSS stands at 0.14% (4th percentile), indicating limited opportunistic exploitation activity despite the broad version range affected.
Certificate validation bypass in libcurl (curl 8.9.1 through 8.21.0) with the wolfSSL TLS backend allows a network-positioned attacker to intercept encrypted communications by exploiting a post-callback CA store reinstatement bug. When an application uses CURLOPT_SSL_CTX_FUNCTION to install a custom trust store at connection time, libcurl silently overwrites that callback-selected store with the previously cached CA bundle, causing certificates that the callback explicitly rejected to be accepted without error. EPSS is 0.13% (3rd percentile) and no active exploitation is confirmed, but exploitation against a susceptible deployment is conceptually straightforward for a MITM-positioned adversary with a suitable certificate.
Memory exhaustion via crafted HTTP stream causes a persistent denial of service on Cisco Desk Phone 9800 Series, IP Phone 7800 and 8800 Series, and Video Phone 8875 running Cisco SIP Software, requiring a manual reboot to recover. The root cause is CWE-401 (missing release of memory after effective lifetime) in the device's HTTP packet processing path: an attacker sends a continuous stream of specially crafted HTTP packets, causing the phone to leak memory until it becomes unavailable. Exploitation is gated on a non-default configuration - Web Access must be explicitly enabled on devices registered to Cisco Unified Communications Manager - materially limiting real-world attack surface despite the network-accessible vector. No public exploit code and no CISA KEV listing were identified at time of analysis.
Unauthenticated sensitive data exposure in dotstore's WooCommerce Product Attachment plugin (versions <= 2.3.3) allows remote, unauthenticated attackers to retrieve product attachment files that should be access-controlled. The vulnerability stems from insufficient authorization checks on attachment retrieval endpoints, classified under CWE-497 (Exposure of Sensitive System Information to an Unauthorized Control Sphere). No active exploitation has been confirmed by CISA KEV, and no public exploit code is identified at time of analysis, but the low attack complexity and zero-authentication requirement make opportunistic access trivial for any network-reachable attacker.
libcurl's SASL state machine for LDAP authentication incorrectly treats an incomplete cryptographic handshake as a fully verified authentication exchange, allowing a Man-in-the-Middle attacker to inject a premature termination message that bypasses peer validation entirely. All curl releases from 7.82.0 through 8.21.0 are affected, spanning a multi-year release history documented in EUVD-2026-72164 and the curl project's own advisory. No public exploit code has been identified and the vulnerability is not listed in CISA KEV, though successful exploitation grants an attacker the ability to intercept LDAP credentials and inject falsified directory responses.
Session fixation in Jenkins 2.579 and earlier (LTS 2.568.2 and earlier) allows an attacker who can serve content on the same site as Jenkins to inject a known session cookie into a victim's browser; when the victim subsequently authenticates via the persistent 'remember me' cookie without session rotation, the attacker inherits the fully authenticated session. The attack chain requires low-privilege same-site content delivery access and active victim participation in a remember-me login flow, meaningfully constraining real-world exploitability. Vendor patches (Jenkins 2.580 and LTS 2.568.3) are available per the Jenkins Security Advisory published 2026-09-02; no public exploit code or CISA KEV listing exists at time of analysis.
Privilege escalation in the Motorola Smart Connect mobile application stems from insufficient authorization controls (CWE-862) on the dashboard UI, which can be manipulated by third-party applications installed on the same device. When paired with a phishing attack targeting the device user, this manipulation enables an attacker to escalate privileges within the device or connected system. No public exploit has been identified at time of analysis, and active exploitation has not been confirmed by CISA KEV.
Path traversal in Kibana's Fleet feature enables a low-privileged attacker with Fleet Settings write access to plant a malicious pathname in settings that, when subsequently acted upon by an administrator, causes deletion of privileged internal resources including user accounts and organizational assets. All Kibana versions prior to 8.19.18, 9.3.6, and 9.4.3 are affected per ESA-2026-158. No public exploit code or active exploitation has been confirmed; the two-party nature of the attack - requiring both attacker write access and administrator interaction - moderates opportunistic exploitability despite the High CVSS rating.
Cross-site scripting (XSS) in the Drupal Screenshot contributed module allows a high-privileged authenticated attacker to inject malicious script content that executes in the browser context of other users who view the affected output, crossing security boundaries as reflected by the scope-changed CVSS vector. All tracked versions of the module are affected per the wildcard CPE and EUVD-2026-70010 listing. No public exploit code and no CISA KEV listing have been identified at time of analysis; Drupal's security team published advisory SA-CONTRIB-2026-102.
Cross-site scripting in the Drupal Screenshot contributed module allows a high-privileged authenticated attacker to inject malicious script content that executes in the browser context of other users who view the affected page. The scope-changing CVSS vector (S:C) confirms that the injected payload can compromise sessions or data belonging to users other than the attacker, enabling session hijacking, credential theft, or unauthorized actions on behalf of victims. No public exploit code has been identified at time of analysis, and no KEV listing is present.
Stored XSS in the Broken Link Checker WordPress plugin (versions ≤ 2.4.13) allows unauthenticated attackers to inject persistent JavaScript into the WordPress admin link log via a multi-hop redirect-based payload delivery mechanism. The plugin's HTTP checker follows attacker-controlled redirects and stores the redirect destination URL verbatim in the link log without sanitization; the admin table printer then renders this unsanitized content to any administrator who views the link log. No public exploit has been identified at time of analysis, and a vendor-released patch is confirmed available as version 2.4.13.1.
Shell command injection in Nuclio's Dashboard build pipeline (versions 1.6.19–1.17.1) allows an authenticated user with function-create permission to execute arbitrary commands inside the Dashboard container by embedding shell metacharacters in the `spec.build.tempDir` field. The post-exploitation impact is substantially amplified because the Dashboard container's Kubernetes service account holds wildcard access to Secrets, Pods, Jobs, and Deployments within its namespace, enabling full namespace-level privilege escalation from a single low-privileged API call. No public exploit identified at time of analysis, but the fix commit (PR #4190) includes working proof-of-concept test cases demonstrating semicolon, pipe, subshell, and backtick injection payloads.
Tencent AI-Infra-Guard's skill-scan component allows scanner bypass via hardcoded exclusion of Python compiled bytecode, enabling malicious skill packages to pass a false SAFE verdict and execute arbitrary code on operator hosts. Attackers supply a skill with benign `.py` source decoys alongside malicious `.pyc/.pyo/.pyd` bytecode in `__pycache__` directories; because three separate scanning surfaces in versions through v4.6.0 unconditionally skip these artifacts, the auditing LLM never sees the malicious payload. CPython's PEP 552 UNCHECKED_HASH behavior executes `.pyc` at import time regardless of source presence, triggering code execution when an operator installs the skill. No public exploit identified at time of analysis, but the upstream fix commit and GitHub Issue #531 confirm the flaw.
GraphQL mutation resolvers in Craft CMS allow authenticated users to create, modify, or delete content entries in sites they are not authorized to access. The `saveEntry` and `deleteEntry` resolvers read `siteId` directly from raw `$arguments` without invoking `ArgumentManager::prepareArguments()`, which is the function that enforces per-site scope filtering via `array_intersect`. Because query resolvers invoke this guard but mutation resolvers do not, the multi-site authorization boundary is structurally absent for all write operations. Vendor-released patch is available in version 5.10.13, and no public exploit has been identified at time of analysis.
Python bytecode scanning gap in agentverus-scanner allows supply-chain attackers to smuggle malicious __pycache__ .pyc files past the tool's static and semantic analysis engines, which only inspect source .py files in companion code directories. Packages carrying malicious bytecode alongside benign source receive a CERTIFIED verdict with elevated trust scores, causing downstream environments that trust the scanner to silently execute attacker-controlled code on import. No public exploit code has been identified at time of analysis, but the VulnCheck advisory and GitHub issue #27 document the bypass technique in sufficient detail for reproduction.
Heap buffer overflow in PX4 Autopilot's sd_bench command allows adjacent unauthenticated attackers to crash the flight controller or potentially execute arbitrary code by supplying a block size parameter smaller than four bytes. The overflow occurs because a fixed four-byte block number is written into a heap allocation sized by the user-supplied block_size value - when that value is 1, 2, or 3 bytes, the write extends beyond the allocated region. No active exploitation has been confirmed (not in CISA KEV) and no EPSS score is available, but the adjacent network attack vector means exploitation is constrained to operators with direct access to the drone's MAVLink control network.
Path traversal via Stapler data binding in Jenkins ThinBackup Plugin 2.1.4 and earlier allows authenticated low-privileged attackers to overwrite the plugin's backup configuration, redirecting backup output to an attacker-specified directory and injecting arbitrary Jenkins controller filesystem files into backup archives. The primary risk is exfiltration of high-value Jenkins controller secrets - credentials.xml, SSH keys, API tokens, master.key - via the backup mechanism. SSVC rates exploitation as none with no public exploit identified; however, the C:H impact makes this a significant post-initial-access escalation path on CI/CD infrastructure.
BEAM atom table exhaustion in Boruta Server prior to 0.10.0 allows any authenticated end user to crash the entire OAuth 2.0/OpenID Connect umbrella application by repeatedly submitting PUT /users/settings requests with arbitrary novel body keys. The UserSettingsController calls String.to_atom/1 on all user-supplied map keys before any validation, permanently interning them in the Erlang VM's global atom table (hard cap: 1,048,576 entries). Once exhausted, the BEAM runtime aborts with 'no more index entries in atom_tab', taking down the auth, admin, and gateway components simultaneously. No public exploit identified at time of analysis, and the issue has been patched in version 0.10.0.
Incorrect authorization in Sonatype Nexus Repository 3 allows a low-privileged user holding the `nexus:blobstores:create` permission to invoke two blobstore group management REST API endpoints that should instead require `nexus:blobstores:update`. By calling these endpoints, the user can convert any existing blobstore into a group blobstore - a structural configuration change - without administrator approval. No public exploit has been identified, but the issue is self-reported by Sonatype and a patch is available in version 3.96.0.
Missing authorization enforcement in Craft CMS's assets/move-asset endpoint allows authenticated low-privilege users to move assets into other users' folders and force-delete conflicting files by supplying the force=1 parameter. Versions 5.0.0 through 5.10.10 are affected; version 5.10.11 resolves the flaw. No public exploit code or active exploitation has been identified at this time, but the low attack complexity and absence of special prerequisites beyond a valid account make exploitation straightforward for any authenticated user.
Unauthenticated cross-site scripting in the Estatik real estate WordPress plugin affects all versions up to and including 4.3.4, allowing remote unauthenticated attackers to inject arbitrary JavaScript into pages viewed by other users. The CVSS scope-change flag (S:C) indicates the injected script can affect browser sessions beyond the vulnerable component itself - typically enabling session hijacking, credential theft, or unauthorized actions performed on behalf of logged-in administrators. No public exploit code or active exploitation via CISA KEV has been identified at time of analysis.
Reflected cross-site scripting in the Login With Ajax WordPress plugin (versions through 4.5.1) allows unauthenticated remote attackers to inject and execute arbitrary JavaScript in a victim's browser by tricking them into clicking a crafted URL. The scope change in the CVSS vector (S:C) reflects the XSS payload executing outside the origin context, enabling session theft, credential harvesting, or DOM manipulation against authenticated WordPress users. No public exploit code or active exploitation has been identified at time of analysis.
Cross-site request forgery in the Activity Log WordPress plugin (versions <= 2.13.1) allows unauthenticated remote attackers to perform unauthorized state-changing actions against the plugin by tricking an authenticated administrator into visiting a malicious page. The vulnerability stems from missing or insufficient nonce validation on plugin requests (CWE-352), and the changed scope (S:C) in the CVSS vector indicates the forged request can affect components beyond the plugin itself - typically the broader WordPress installation. No public exploit code or active exploitation has been identified at time of analysis.
Incomplete file coverage in claude-skill-antivirus v2.1.3 allows malicious Claude skill packages to bypass security scanning entirely when installed from local directories. The scanner reads only the SKILL.md manifest file, leaving Python source (.py), bytecode (.pyc), and all other executable artifacts in the scripts/ subdirectory completely unanalyzed. Skill authors can embed arbitrary malicious payloads in non-manifest files and receive a SAFE verdict with a perfect 100/100 trust score, providing users with false assurance before installation.
Unauthenticated cross-site scripting in the TrustedSite WordPress plugin (versions <=1.2.5) allows remote attackers without any credentials to inject arbitrary JavaScript into pages served to site visitors or administrators. The CVSS vector designates a scope change (S:C), meaning the injected script executes in the victim's browser context and can affect resources beyond the vulnerable WordPress installation itself - including session cookies and browser-stored credentials. No public exploit code or CISA KEV listing has been identified at time of analysis, but the unauthenticated entry point and low attack complexity make this a realistic risk on any unpatched deployment.