Wildcard DNS SAN matching in python-cryptography's X.509 chain verifier prior to 49.0.0 allows a leaf certificate bearing *.example.com to satisfy a constrained intermediate CA's permittedSubtrees restriction of foo.example.com, enabling scope escape to sibling domains such as bar.example.com. Applications relying on the PolicyBuilder/Store verification API are exposed to acceptance of invalid certificate chains that should be rejected per RFC 5280 §4.2.1.10. A public proof-of-concept is confirmed in the GHSA advisory; the vulnerability is not in the CISA KEV catalog, but the CVSS 4.0 E:P modifier reflects the POC's availability.
Channel lifecycle enforcement failure in russh (Rust SSH library, versions <= 0.62.4) allows authenticated clients to invoke server-side handler callbacks - including exec_request, shell_request, subsystem_request, data, channel_eof, and channel_close - for SSH channel IDs that were never opened or confirmed by the transport layer. An authenticated peer can send crafted SSH_MSG_CHANNEL_REQUEST packets without ever issuing SSH_MSG_CHANNEL_OPEN, or after having a channel_open_session request explicitly denied by the application, and still trigger all channel-scoped callbacks. A working Python PoC is publicly available demonstrating 32 exec_request invocations with zero channels ever opened; no active exploitation is confirmed (not in CISA KEV) and a vendor patch exists in v0.62.5.
Cookie domain scope bypass in Guzzle PHP HTTP client (prior to 7.15.2 and 8.0.1) allows improper cookie sharing via non-canonical IP address representations in cookie Domain attributes. The flaw in SetCookie::matchesDomain() fails to recognize hexadecimal (e.g., 0x7f000001), mixed-base octal (e.g., 0177.0.0.0x1), and percent-escaped (e.g., 192.168.0.%31) host spellings as IP literals, so cookies set for such domains incorrectly gain subdomain scope while libcurl resolves the same strings as actual IP addresses. This divergence enables an adversary-controlled host (e.g., evil.0x7f000001) to receive session cookies intended for the target IP, facilitating credential disclosure and session fixation; no public exploit identified at time of analysis.
Unauthorized lesson content disclosure in Academy LMS WordPress plugin before 3.8.3 allows any subscriber-level authenticated user to read arbitrary lesson content via the REST API - including lessons from paid courses they have not purchased and lessons in unpublished states (draft, pending, private). The root cause is a complete absence of enrollment and publication-status validation in the single-lesson REST API endpoint. A publicly available proof-of-concept exists per WPScan, though EPSS at 0.13% (3rd percentile) and SSVC classification as non-automatable indicate limited observed exploitation activity at the time of analysis.
Post-deletion authorization bypass in the Contest Gallery WordPress plugin before 30.0.7 allows any authenticated Author-level or higher user to permanently delete arbitrary posts, pages, and other WordPress content they do not own. The plugin's post-deletion AJAX handler enforces only a coarse role-membership test, entirely omitting per-object capability verification and nonce validation - the two controls that WordPress's built-in permission model relies on to prevent cross-user destructive actions. A public exploit is available via WPScan, lowering the skill barrier; though EPSS is low at 0.13%, any multi-author site with untrusted Author accounts faces real and irreversible content destruction risk.
Broken access control in Simply Schedule Appointments WordPress plugin before 1.6.12.11 exposes all customer appointment records to any authenticated Contributor-level user site-wide. The administrative appointment-listing shortcode omits a capability check, and the per-user result scoping logic defaults to showing all records when the requesting user is not recognized as staff. Any WordPress user holding the Contributor role or higher - a commonly granted role in multi-author sites - can enumerate names, email addresses, phone numbers, and appointment notes belonging to every customer in the system. A publicly available proof-of-concept exists, though EPSS at 0.13% and SSVC Automatable:No suggest exploitation remains targeted rather than widespread.
Uncontrolled search path exploitation in CheckMAL AppCheck Pro 3.1.43.10's kernel mini-filter driver (AppCheckD.sys) allows a local attacker with low privileges to redirect the driver's library loading path, potentially achieving kernel-level code execution on the target host. A public proof-of-concept exploit has been released despite the vendor declining to respond to coordinated disclosure, making this a no-patch, public-exploit scenario. While high attack complexity (AC:H) reduces opportunistic risk, the complete confidentiality, integrity, and availability impact at the kernel layer makes successful exploitation highly damaging.
Path traversal in PostCSS npm package (versions through 8.5.22) allows disclosure of arbitrary `.map` files from the server filesystem when applications process attacker-controlled CSS without the `from` option. The vulnerability is an incomplete fix bypass of prior advisory GHSA-6g55-p6wh-862q: the traversal guard added to `PreviousMap.loadFile()` was gated on `cssFile` being truthy, which is never the case when `opts.from` is absent. No CISA KEV listing, but a verified proof-of-concept is included in the upstream GHSA advisory GHSA-fxqj-rqcc-2cmp. The CVSS 4.0 score of 6.3 with AT:P correctly reflects the specific deployment condition required.
HTTP request smuggling in aiohttp's server-side parser allows unauthenticated remote attackers to inject and misdirect HTTP traffic by exploiting a protocol-switch race in the WebSocket upgrade path. Versions 3.14.1 and earlier prematurely mark the connection as upgraded before the full request body is consumed, causing trailing bytes - potentially an attacker-controlled HTTP request - to be interpreted as WebSocket or pipelined data by the server or any upstream proxy. No public exploit code exists and the vulnerability is not listed in CISA KEV, but CVSS 4.0 AC:H reflects the edge-case nature of the trigger condition.
SQL injection in Adobe Campaign Classic (ACC) lets a remote, unauthenticated attacker inject arbitrary SQL commands that escalate to arbitrary code execution in the context of the current user, per Adobe advisory APSB26-120. The CVSS 3.1 base score is a maximum 10.0 with a changed scope, reflecting network reachability, no privileges, no user interaction, and full confidentiality/integrity/availability impact spilling into adjacent components. There is no public exploit identified at time of analysis and it is not listed in CISA KEV, but the maximum severity and unauthenticated network vector make this a top-priority patch.
Stored cross-site scripting in the Simple Membership WordPress plugin before 4.7.8 allows unauthenticated attackers to inject arbitrary JavaScript into the WordPress administration dashboard by supplying a malicious subscriber name in a payment approval request. Because the plugin neither sanitizes the input at ingestion nor escapes it on output when rendered for administrators, the payload persists server-side and executes in any administrator's browser session upon viewing the affected dashboard panel. A public proof-of-concept exploit exists (reported by WPScan); however, EPSS stands at 0.17% (7th percentile) and CISA KEV listing is absent, indicating limited observed mass exploitation at time of analysis.
Stored cross-site scripting in the Blog Floating Button WordPress plugin (≤1.4.20) allows unauthenticated attackers to inject malicious JavaScript via the HTTP User-Agent header into the plugin's visitor tracking REST endpoint. The payload persists in the database and executes in the browser session of any administrator who opens the plugin's access report page, enabling full site takeover through the hijacked admin session. A publicly available proof-of-concept from WPScan confirms the attack path; no vendor-released patch has been identified at time of analysis.
Stored XSS in the SVG Support WordPress plugin (before 2.5.17) is achieved by uploading a script-bearing .svgz file, bypassing the plugin's sanitization logic that applies only to .svg uploads despite both extensions being registered and served as SVG content. Any WordPress user role granted SVG upload access - such as an Author - can plant the payload, which then executes in the browser of any subsequent visitor, including administrators, enabling session hijacking or privileged action injection. A public proof-of-concept exists via WPScan; EPSS is low (0.16%, 6th percentile) and SSVC rates exploitation as non-automatable, but the stored nature of the payload and admin-reachability elevate practical impact.
Remote code execution in Adobe Campaign Classic (ACC) arises from server-side template injection (CWE-1336), where improperly neutralized special elements in the template engine let an attacker run arbitrary code in the context of the current user. Any exposed ACC instance is affected, and per the CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N) exploitation is remote, low-complexity, and requires no privileges or user interaction, with a changed scope pushing the CVSS base score to a maximum 10.0. There is no public exploit identified at time of analysis and the issue is not listed in CISA KEV, but the perfect score and SSTI class make it a high-priority patch target.
Authentication bypass in Tranquil IT Wapt Server before 2.6.1.17813 lets a remote unauthenticated attacker send a specially crafted packet to obtain a valid session token for a targeted account, yielding full account takeover of the endpoint-deployment console. Because Wapt Server orchestrates software installation across managed endpoints, a stolen operator/admin session effectively grants control over the fleet it manages. Rated CVSS 4.0 10.0 with subsequent-system impact; no public exploit identified at time of analysis and it is not listed in CISA KEV.
Server-Side Request Forgery in Adobe Campaign Classic (ACC) enables remote attackers to coerce the server into issuing crafted internal requests, resulting in privilege escalation without any user interaction. Adobe rates the flaw at the maximum CVSS 10.0 with a changed scope, indicating the SSRF pivots beyond the vulnerable component to compromise other systems or trust boundaries. There is no public exploit identified at time of analysis and it is not listed in CISA KEV, but the unauthenticated, network-reachable nature makes it a top-priority patch.
SQL injection in Adobe Campaign Classic (ACC) allows a low-privileged, authenticated attacker to execute arbitrary code in the context of the current user without any user interaction. Because unsanitized input reaches an SQL command (CWE-89), an attacker with minimal access can pivot from database manipulation to code execution, and the CVSS scope change (S:C) indicates the impact extends beyond the vulnerable component itself. No public exploit identified at time of analysis, but the 9.9 base score and Adobe's own advisory (APSB26-120) mark this as a top-priority enterprise marketing-platform flaw.
SQL injection in SiYuan's fullTextSearchAssetContent endpoint lets unauthenticated attackers and holders of publish RoleReader tokens execute arbitrary SQL against the read-write asset-content database in all versions before v3.7.3. Because the endpoint is reachable without authentication and the database is writable, attackers can read, alter, or delete data across notebooks - not merely exfiltrate it. No public exploit identified at time of analysis, but the flaw was reported by VulnCheck and carries a CVSS 4.0 base score of 9.9.
SQL injection in SiYuan personal knowledge-management software (all versions before v3.7.3) lets attackers inject arbitrary SQL through the unescaped keyword parameter of the /api/filetree/searchDocs endpoint. Because the query runs on a read-write SQLite handle that permits stacked statements, an attacker can both read and tamper with content across every cleartext notebook, and the endpoint is reachable unauthenticated when publish mode runs with Publish.Auth.Enable=false. Reported by VulnCheck with a vendor patch available; no public exploit identified at time of analysis, and the CVSS 4.0 score is 9.9.
Root-level OS command injection in the openmediavault-md plugin of OpenMediaVault v8.0.4-1 lets attackers inject shell metacharacters into a parameter that is passed unsanitized to a system command, yielding arbitrary command execution as root. OpenMediaVault is a Debian-based network-attached-storage (NAS) management platform, and the -md plugin manages Linux mdadm software RAID arrays. Publicly available exploit code exists (a proof-of-concept gist is referenced), though there is no public exploit identified in CISA KEV; EPSS is modest at 0.62% (46th percentile).
OS command injection in the Serverless-Devs CLI (@serverless-devs/s versions 3.1.11 and earlier) lets an attacker execute arbitrary commands on a developer's machine when they run 's init' with a crafted, attacker-influenced argument. The tool passes unsanitized input into child_process.spawn() with shell:true, and the single guard check - requiring the value to end in '.git' - is trivially satisfied by an attacker while still smuggling shell metacharacters. Publicly available exploit code exists (a proof-of-concept gist), SSVC rates technical impact as total and automatable, but EPSS remains low at 0.48% and it is not in CISA KEV.
Privilege escalation in Adobe Campaign Classic (ACC) stems from an incorrect authorization flaw (CWE-863) that lets an attacker gain elevated privileges without any user interaction. The CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N) scores it 9.8 Critical, implying a network-reachable, low-complexity path to full compromise of confidentiality, integrity, and availability. There is no public exploit identified at time of analysis and the CVE is not listed in CISA KEV, so exploitation is currently theoretical based on the vendor advisory.
Predictable API key generation in osTicket 1.18.3 lets remote attackers reconstruct valid staff API keys by exploiting weak MD5-based construction seeded with the request timestamp and client IP address (CWE-331). Because the key is derived from low-entropy, guessable inputs, an attacker who approximates the key creation window can brute-force the key space and gain API access without credentials. Rated CVSS 9.8 by the reporter; no public exploit is identified at time of analysis, though a technical disclosure write-up exists, and CISA SSVC lists exploitation as none but marks the flaw automatable with total technical impact.
Authorization bypass in Menulux Software Inc. Mobile App (all versions through the 12.05.2026 build) lets remote attackers manipulate a user-controlled key (an Insecure Direct Object Reference, CWE-639) to reach or alter resources belonging to other users, enabling a software integrity attack. TR-CERT rates it CVSS 9.8 with full CIA impact, though the primary documented effect is integrity tampering. There is no public exploit identified at time of analysis and the CVE is not listed in CISA KEV.
Authentication bypass in the Net::SAML2 Perl module before 0.86 lets an unauthenticated attacker log in as any user against Service Providers that accept EncryptedAssertions. The flaw stems from _verify_encrypted_assertion returning a decrypted assertion as 'verified' whenever it contains no dsig:Signature element, skipping both the signature and trust-anchor checks. Because an SP publishes its encryption certificate in SAML metadata, anyone can encrypt a forged unsigned assertion to it, wrap it in a samlp:Response, and post it to the assertion consumer service; no public exploit identified at time of analysis but the logic bug is trivially reproducible.
SQL injection in FastAdmin v1.6.1.20250430 lets remote attackers inject arbitrary SQL through the application/common/controller/Backend.php component, with the vendor advisory characterizing the impact as extending to arbitrary code execution. The flaw is tagged RCE/SQLi/PHP and carries a CVSS 9.8 rating with total technical impact per CISA's SSVC framework, though publicly available exploit code exists rather than confirmed active exploitation. EPSS is low (0.14%, 4th percentile), suggesting exploitation is not yet widespread despite the critical severity.
Remote code execution in Microsoft Edge (Chromium-based) before 151.0.4129.59 allows a network attacker to run arbitrary code by exploiting a type-confusion condition (CWE-843) when a victim is lured to malicious web content. Because the CVSS scope is Changed (S:C), successful exploitation can escape the affected component and impact the broader system, making this a high-severity browser flaw. There is no public exploit identified at time of analysis and SSVC reports exploitation as 'none', but the total technical impact and 9.6 CVSS score warrant prompt patching.
Remote code execution in Adobe Campaign Classic (ACC) lets an authenticated low-privileged attacker inject directives into dynamically evaluated code (CWE-95 eval injection) and run arbitrary code in the context of the current user. Because the CVSS scope is changed, the impact can reach components beyond the initially exploited one, driving the 9.6 critical score. Reported by Adobe with advisory APSB26-120; no public exploit identified at time of analysis and no CISA KEV listing.
Path traversal in danpros HTMLy up to version 3.1.1 allows remote unauthenticated attackers to read arbitrary files from the server filesystem by manipulating the Author Name argument in /system/htmly.php. A publicly available proof-of-concept exploit (E:P per CVSS 4.0 supplemental metric) exists at a referenced GitHub issue, materially lowering the exploitation bar. The vendor was unresponsive to responsible disclosure, leaving no vendor-confirmed patch and no official mitigation guidance.
Out-of-bounds read in mz-automation libiec61850 up to version 1.6.1 allows remote unauthenticated attackers to crash the IEC 61850 MMS server process and potentially leak heap or stack memory contents by sending a crafted MMS DataSet access request. The root cause is unconstrained string copy operations in the checkDataSetAccess function, now confirmed fixed in v1.6.2 via bounded copy with a 129-byte limit. A public exploit has been disclosed; however, this vulnerability is not listed in CISA KEV and the CVSS 4.0 score of 6.9 (Medium) reflects limited impact scope confined to availability of the vulnerable system.
Remote unauthenticated crash in mz-automation libiec61850 1.6.0-1.6.1 allows a network attacker to terminate an IEC 61850 MMS server by supplying an oversized Report ID string to the Report Control Block write handler, triggering a free of non-heap memory (CWE-590). No authentication or user interaction is required per the CVSS 4.0 vector (PR:N/UI:N/AV:N), and a public proof-of-concept exploit exists at the referenced GitHub issue. Active exploitation has not been confirmed by CISA KEV, but the combination of a public PoC and trivial exploitation path elevates practical risk in operational technology environments where IEC 61850 server continuity is safety-critical.
Unauthenticated remote access to the /EIP_Com_FileList.aspx endpoint in NewType WebEIP up to version 3.0 exposes file listing functionality to any network-reachable attacker due to improper authentication enforcement (CWE-287). A public proof-of-concept exploit hosted on Google Drive is confirmed available, lowering the exploitation bar significantly. No vendor patch exists as the vendor did not respond to disclosure; organizations must rely entirely on compensating controls.
Server-side request forgery in jina-ai Reader's Puppeteer-based crawler allows unauthenticated remote attackers to manipulate the `isValidTLD` function to coerce the server into issuing arbitrary HTTP requests to internal or restricted network targets. All deployments up to commit 1574bfd380d249c86c82db4dace0d9c8fe17e2b1 are affected with no vendor-issued patch, as the maintainers did not respond to disclosure. A public proof-of-concept exploit exists (no CISA KEV listing), and the rolling release model means there is no safe pinned version for operators to target.
Authentication bypass in Check Point Security Management Server and Multi-Domain Security Management Server (MDS) lets an unauthenticated remote attacker with network access to Management services execute arbitrary commands and fully compromise the management system. It affects a broad range of releases (R80 through R82.10) including R81.20 and R82 running Jumbo Hotfix Accumulator takes at or below fixed thresholds. There is no public exploit identified at time of analysis; Check Point found the flaw internally and reports no indication of active exploitation despite the critical CVSS 4.0 score of 9.3.
Stack-based buffer overflow in the Wavlink WL-NU516U1 (firmware 708c073-mt7628) router's nas.cgi handler lets remote unauthenticated attackers overflow a fixed-size stack buffer by manipulating the HTTP CONTENT_LENGTH value processed via fgets. Successful exploitation can crash the device or execute arbitrary code with full compromise of confidentiality, integrity, and availability. A public write-up (GitHub report2.md) exists, but the flaw is not listed in CISA KEV - no public weaponized exploit is confirmed at time of analysis.
Remote code execution in Tenable Sensor Proxy (versions before 1.4.2) lets an attacker run code with elevated privileges by luring an operator into connecting the sensor to an attacker-controlled host. The CWE-94 code-injection flaw is triggered during that connection, and the CVSS 4.0 score of 9.3 reflects full compromise of confidentiality, integrity, and availability on the proxy host. There is no public exploit identified at time of analysis and it is not on CISA KEV, but a vendor patch is available.
Certificate name-constraint bypass in Bouncy Castle for Java allows a crafted certificate to escape X.509 permitted/excluded subtree restrictions by appending a trailing dot to an rfc822Name (email) or URI host, or by prepending RDNs ahead of a permitted directoryName. Any relying party using BC's PKIXNameConstraintValidator (via CertPathValidator/CertPathBuilder) to enforce name constraints can be tricked into trusting an identity a constrained CA was never authorized to issue, enabling impersonation and effective authentication bypass. There is no public exploit identified at time of analysis, but the vendor rates it 9.3 (Critical) and a fix is available.
TLS hostname-verification bypass in the Bouncy Castle for Java JSSE provider (BCJSSE) ships the legacy certificate-CN fallback enabled by default, even though the vendor documentation states it is opt-in and gated behind the org.bouncycastle.jsse.client.checkCnFallback property. Affected are BC-Java before 1.85, BC-LTS before 2.73.12, and BC-FJA (FIPS) TLS module bctls-fips before 1.0.24, 2.0.24 and 2.1.24; a MITM-positioned attacker can present a SAN-less certificate whose Subject CN names the victim host and have it silently accepted, defeating name-constraint protections and enabling interception of TLS traffic (CWE-297). This carries a CVSS 4.0 base of 9.3, but there is no public exploit identified at time of analysis and it is not in CISA KEV.
Private key recovery in Bouncy Castle for Java (bc-java before 1.85, bc-lts-java before 2.73.12) affects the DH agreement code path, where a peer-supplied public value is exponentiated with the local private key without full range/subgroup validation. A malicious protocol peer can submit small-order or out-of-range Diffie-Hellman elements to mount a small-subgroup confinement attack and, when the local private key is static/reused, gradually recover it via CRT - enabling decryption and impersonation. No public exploit identified at time of analysis; the fix is a vendor-released patch and the vulnerability is not listed in CISA KEV.
Certificate revocation-check bypass in the Bouncy Castle for Java (bc-java) PKIX provider allows a validly-signed OCSP response issued for one certificate to be accepted as proof-of-status for a different certificate, because the stapled response is not bound to the CertID of the certificate actually being checked. This means a revoked end-entity certificate can pass PKIXRevocationChecker validation when a 'good' OCSP response for another serial (e.g. the CA's) is supplied, defeating revocation enforcement in TLS/PKI clients that rely on OCSP stapling. Affected releases are bc-java before 1.85, bc-lts-java before 2.73.12, and BC-FJA (FIPS) before 2.0.2 and 2.1.3; no public exploit identified at time of analysis and it is not listed in CISA KEV.
Authentication bypass in SiYuan (self-hosted knowledge-management/note-taking app) versions before v3.7.3 allows anonymous readers in publish mode to exfiltrate the full content of password-protected documents. While the primary getDoc endpoint enforces the publish-mode password, sibling content-returning endpoints (getHeadingChildrenDOM, the getHeading*Transaction family, and getBacklinkDoc) omit that check, so an attacker who harvests internal block IDs from reader-accessible endpoints can pull protected content directly. VulnCheck reported the issue and a vendor patch (v3.7.3) is available; there is no public exploit identified at time of analysis and it is not listed in CISA KEV.
Information disclosure in SiYuan (a personal knowledge-management/note-taking application) before v3.7.3 lets anonymous readers or low-privilege publish RoleReader tokens read the full rendered content of documents that were explicitly marked publish-disabled. The getHeadingDeleteTransaction, getHeadingLevelTransaction, and getHeadingInsertTransaction endpoints return rendered block DOM without enforcing publish-access checks, so supplying a target heading block ID leaks restricted content. Reported by VulnCheck with a vendor patch available; no public exploit identified at time of analysis, and this is not listed in CISA KEV.
Broken access control in SiYuan versions before 3.7.3 lets publish-mode readers - including anonymous users when publish Basic Auth is disabled - retrieve the rendered content of publish-forbidden documents by calling the /api/ref/getBacklinkDoc and /api/ref/getBackmentionDoc endpoints directly with a target document ID. Although the corresponding backlink list endpoints correctly filter publish-forbidden documents, these two content endpoints are gated only by CheckAuth and omit the publish-access filter, exposing DOM content and a reference-existence oracle. Reported by VulnCheck; no public exploit identified at time of analysis and not listed in CISA KEV.
Signature validation bypass in Misskey (versions 12.37.0 through prior to 2026.5.4) allows an attacker to have spoofed ActivityPub activities accepted as authentic. The flaw resides in the JSON-LD signature validation and compaction logic (CWE-347), enabling forgery of federated actions such as posts or actions attributed to other actors/instances. The CVSS 4.0 base score is 9.2 (Critical) with a network, unauthenticated, low-complexity vector emphasizing integrity impact; no public exploit has been identified at time of analysis, and it is not listed in CISA KEV.
Signature-verification bypass in the Perl XML::Sig module before 0.71 allows remote attackers to forge which document node is validated by injecting XPath operators through the SignedInfo/Reference/@URI attribute. Because verify() and _get_signed_xml() concatenate the attacker-controlled URI into an XPath expression without escaping or NCName validation, a single quote breaks out of the string literal and lets the injected expression decide which element is digest-checked - undermining the integrity guarantee of any XML Signature the library verifies. No public exploit has been identified at time of analysis, and the flaw is not on CISA KEV; a vendor patch exists in release 0.71.
Authentication and authorization bypass in @fastify/aws-lambda 6.4.0 lets an unauthenticated remote attacker forge the entire AWS Lambda proxy event by setting a single client-controlled HTTP header. Because the request.awsLambda.event/context decoration reads the untrusted x-apigateway-event and x-apigateway-context headers before falling back to the trusted internal invocation token, any application that consults these values for identity or API Gateway authorizer claims can be tricked into granting attacker-chosen access, enabling full privilege escalation. Rated CVSS 9.1; no public exploit identified at time of analysis, but the attack is trivial and documented in the vendor advisory.
Signature verification bypass in the Perl XML::Sig module before 0.71 allows XML Signature Wrapping (XSW) via duplicate ID attributes, letting attackers forge signed content that still validates. verify() resolves a signed Reference URI with the XPath "//*[@ID='$id']" and trusts the first matching node, so a document containing two elements with the same ID passes verification while a downstream consumer reads the attacker-supplied second element; in a SAML2 SP this puts Assertion contents under attacker control. Rated CVSS 9.1 (network, no authentication, no user interaction); no public exploit identified at time of analysis, though the vendor's fix ships a regression test that reproduces the technique.
Man-in-the-middle credential theft and AI-response injection affects Emlog Pro through 2.6.23, where the AI integration module (include/service/ai.php) unconditionally disables TLS certificate validation on every outbound HTTPS call to configured LLM providers. A network-adjacent attacker who can intercept this traffic can present any TLS certificate to harvest the Authorization Bearer API keys sent with each AI request and to forge LLM responses that feed the automated tool-call pipeline, including the query_database and update_config handlers. There is no public exploit identified at time of analysis and it is not in CISA KEV, but the CVSS 4.0 base score of 9.1 reflects the high confidentiality and integrity impact.
OS command injection in OpenWrt's luci-app-dockerman grants any authenticated user holding only the read ACL full root-level code execution on the affected router. The vulnerable run_ttyd handler in the ucode docker_rpc.uc backend constructs a shell command from attacker-controlled request fields (id, cmd, uid) and passes them unsanitized to system() within the rpcd root context, reachable via a simple HTTP POST to /ubus. Only LuCI master and openwrt-25.12 snapshots after the JS-to-ucode backend conversion are affected; upstream patch commits exist but no released patched snapshot version was confirmed as of the advisory, and no public exploit is identified at time of analysis.
Stack-based buffer overflow in the Wavlink WL-NU516U1 NAS device's nas.cgi password change handler allows remote unauthenticated attackers to potentially execute arbitrary code by supplying an oversized User1Passwd argument to the change_password function. Affecting the MT7628 firmware variant (708c073-mt7628), this carries a CVSS 4.0 score of 8.9 with full confidentiality, integrity, and availability impact and no privileges or user interaction required. A public proof-of-concept has been published and a vendor-released firmware patch is available as of 2026-07-13; no confirmed CISA KEV listing at time of analysis.