OS command injection in esengine DeepSeek-Reasonix (Reasonix Studio) allows a local attacker who controls a repository's contents to execute arbitrary commands with the privileges of the desktop application's diff viewer. The attacker supplies a crafted .gitattributes file that registers a clean/smudge filter plus a .git/config that binds that filter driver to an arbitrary shell command; when the victim opens that repository's workspace-changes diff view, the app's internal/gitcmd module runs git diff and the malicious filter is executed. Exploitation is strictly local (AV:L) and requires user interaction - the victim must open or view the attacker-supplied repository - and no public exploit code or confirmed active exploitation has been identified at time of analysis, so the risk is genuine but bounded rather than a mass-exploitable network flaw.
Unauthenticated read access to CI/CD job traces in GitLab CE/EE can expose sensitive CI/CD variable values because the GraphQL API fails to enforce proper authorization for trace retrieval, affecting all versions from 15.11 before 19.2.7, 19.3 before 19.3.3, and 19.4 before 19.4.1. Publicly available exploit code exists, with details disclosed through HackerOne report 3590929, and the bug was reported by GitLab itself; no confirmed active exploitation (CISA KEV) has been recorded. The CVSS score is only 3.7 (CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:N/A:N), reflecting high attack complexity and a confidentiality-only impact limited to reading trace contents, so this is a real but modest issue that manifests only under specific, non-universal conditions rather than an urgent, universally exploitable flaw.
Progress Software Fiddler Everywhere 8.0.2 ships a local .NET backend (Fiddler.WebUi) that exposes an HTTP and SignalR RPC endpoint on the loopback interface without any authentication, allowing any local user or process on the same machine to call privileged methods, mint OAuth tokens, and export the application's machine-in-the-middle root certificate. The issue is strictly local (CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N, 7.7) and only exploitable while Fiddler Everywhere is installed and running, but it requires no credentials or user interaction once those conditions hold. There is no evidence of active exploitation (not in CISA KEV) and no public exploit code identified at time of analysis; the practical risk is credential and trust-anchor theft on shared workstations or hosts where untrusted local software can execute.
Certificate validation bypass in urllib3 1.26.0 through 2.7.x allows a man-in-the-middle attacker who can impersonate the configured HTTPS proxy to intercept, observe, or modify traffic forwarded through that proxy, and in some configurations to obtain the target TLS client certificate. The flaw only affects applications that route requests through an HTTPS proxy while also setting target-server TLS options - proxy_ssl_context, proxy_assert_hostname, proxy_assert_fingerprint, ssl_context, cert_reqs, verify_mode, or CERT_NONE - because those target settings are incorrectly applied to the proxy TLS handshake; in particular, cert_reqs=CERT_NONE can overwrite proxy_ssl_context.verify_mode in place and the mutation persists across reused connections, silently disabling proxy certificate verification. End-to-end target TLS via CONNECT tunneling remains protected, and no public exploit identified at time of analysis; the issue is fixed in urllib3 2.8.0.
Engine.IO versions 6.6.0 through 6.6.9 (the transport layer underlying Socket.IO servers on Node.js) can be crashed by a remote, unauthenticated attacker who first completes a normal handshake and then sends a transport upgrade request carrying an EIO query parameter that differs from, or omits, the revision negotiated for that session. The mismatched revision causes the server to attach a transport with an inconsistent parser and heartbeat configuration, and a crafted heartbeat packet can raise an uncaught exception that terminates the Node.js process - a pure denial-of-service condition (CVSS 3.1: 7.5, AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H, CWE-248). No public exploit identified at time of analysis and no CISA KEV listing; risk is nonetheless genuine because the vulnerable transport-upgrade path is enabled by default and the trigger is trivial once a session id is obtained. Vendor-released patch: engine.io 6.6.10.
OpenSSL 3.4.x, 3.5.x, 3.6.x and 4.0.x servers that enable the native QUIC stack can be driven into denial of service through unbounded memory consumption in QUIC STREAM fragment metadata handling (CVE-2026-54873, CWE-770). Remote, unauthenticated attackers (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H, 7.5) can send crafted QUIC streams that cause excessive allocation and service degradation or exhaustion; confidentiality and integrity are unaffected. Exploitation is assessed as not observed (SSVC: Exploitation none; no CISA KEV listing) and no public exploit code has been identified at time of analysis, but the attack is automatable. Risk is materially bounded by the relatively narrow deployment base of OpenSSL's QUIC server functionality - classic TLS termination endpoints are not exposed.
Denial of service in OpenSSL's QUIC stack affects OpenSSL 3.4.0 before 3.4.8, 3.5.0 before 3.5.9, 3.6.0 before 3.6.5, and 4.0.0 before 4.0.3, where an unbounded backlog of RETIRE_CONNECTION_ID frames lets remote unauthenticated attackers exhaust memory by flooding a reachable QUIC endpoint. The flaw is an allocation-without-limits issue (CWE-770) rated CVSS 7.5 (availability-only, AV:N/AC:L/PR:N/UI:N) and addressed in the OpenSSL 4.0.3, 3.4.8, 3.5.9, and 3.6.5 patch releases. No public exploit identified at time of analysis; the attack is automatable, but exposure is limited to deployments that enable OpenSSL's relatively new QUIC support.
Signed integer overflow in the s2 dictionary API of klauspost/compress before 1.18.7 lets an attacker crash a consuming Go process: a dictionary carrying a uvarint-encoded repeat value larger than MaxInt64 slips past repeat-index validation, and a subsequent call to Dict.Encode() converts the overflowed negative repeat value into an out-of-bounds unsafe.Pointer access that aborts the process with SIGSEGV. The issue is availability-only (C:N/I:N, A:H) on a CVSS:3.1/AV:N/AC:L/PR:N/UI:N vector, but real exposure hinges on the application - it only matters where untrusted, attacker-influenced bytes reach s2.NewDict() and the resulting dictionary is then encoded, not in ordinary compression or decompression of arbitrary data. No public exploit code and no CISA KEV entry were identified at time of analysis; the vendor-released patch is v1.18.7.
Arbitrary code execution in WatchGuard Fireware OS is possible through its SAML single sign-on session daemon (samld), which deserializes session files it is asked to load without adequately validating their contents. The attacker must already be authenticated with high privileges (CVSS 4.0 vector PR:H) and must already be able to write files on the appliance, since the vector's AT:P metric and the description both indicate a required prior condition - this is a second-stage, post-foothold vulnerability rather than a standalone remote compromise. Exploitation yields code execution in the security context of the samld service, with high confidentiality, integrity and availability impact to that service but no demonstrated impact on subsequent systems (SC:N/SI:N/SA:N); no public exploit code and no confirmed active exploitation were identified at time of analysis.
Sandbox escape in Google Chrome's WebView component on Android (versions prior to 154.0.8037.92) lets an attacker who has already compromised the renderer process execute arbitrary code outside the browser sandbox by serving a crafted HTML page to the victim. The flaw is a second-stage, chain-completing bug rather than a standalone remote-compromise vector: it requires an existing renderer-level foothold, high attack complexity, and victim user interaction, which is why CISA's SSVC assessment marks exploitation as 'none', automatability as 'no', and Chromium rates its own security severity as Low despite a 7.5 CVSS base score. No confirmed active exploitation (CISA KEV) and no public exploit code were identified at time of analysis; a vendor patch is available in Chrome 154.0.8037.92.
Out-of-bounds memory access in OpenSSL during an in-progress TLS/DTLS handshake can crash or destabilize servers that call SSL_set_SSL_CTX() mid-handshake, producing a denial of service (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H, 7.5). The issue affects OpenSSL 4.0.0 through 4.0.2, 3.6.0 through 3.6.4, 3.5.0 through 3.5.8, and 3.4.0 through 3.4.7, and is fixed in 4.0.3, 3.6.5, 3.5.9, and 3.4.8 respectively. Exploitation is unauthenticated and requires only network reach to a server that swaps SSL_CTX objects inside the SNI (servername) callback for virtual hosting or multi-certificate selection; no public exploit identified at time of analysis, and this is a conditional availability issue rather than a critical priority.
Improper validation in the Russh Rust SSH library (versions prior to 0.63.1) lets a malicious or compromised SSH server drive client-side channel lifecycle callbacks for channel identifiers the client never opened, never confirmed, or already released. The affected code path is client_read_authenticated in russh/src/client/encrypted.rs, which forwards CHANNEL_DATA, CHANNEL_EXTENDED_DATA, CHANNEL_EOF, CHANNEL_CLOSE, CHANNEL_OPEN_FAILURE, CHANNEL_SUCCESS, CHANNEL_FAILURE, and CHANNEL_REQUEST subtypes (exit-status, exit-signal, xon-xoff) to public client::Handler callbacks without checking that the ChannelId belongs to a channel the client opened and established. Exploitation requires the russh-based application to act as an SSH client and connect to a server the attacker controls or has compromised - so risk concentrates in untrusted jump hosts, MITM positions, and automation that reaches attacker-influenced endpoints rather than arbitrary internet-facing victims; no public exploit code has been identified at time of analysis. Successful abuse can panic the application or corrupt command-completion and exit-code tracking, and the vendor assigns CVSS 7.5 while the independent assessment scores integrity and availability impact at High.
Use-after-free in OpenSSL 4.0.0 through 4.0.2's X.509 extension cache allows unauthenticated remote attackers to crash multithreaded services that parse or validate attacker-influenced certificates, producing denial of service. Exploitation is not a deterministic single-shot input: it requires winning a timing-sensitive thread race while the same certificate object's extension cache is accessed concurrently, so single-threaded or non-shared certificate handling is unaffected. NVD lists the issue at CVSS 7.5, while our assessment applies AC:H and confirms availability-only impact (AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H); no public exploit code or KEV-listed active exploitation was identified at time of analysis.
A race condition in Mozilla Firefox's Audio/Video component can culminate in a use-after-free, potentially allowing memory corruption when a victim loads attacker-controlled media content in any Firefox build prior to version 157. Exploitation is unauthenticated over the network but demands both user interaction (the victim must visit or interact with a crafted page) and a precisely won, non-deterministic timing window between concurrent media-processing operations, which makes the bug high-complexity and not automatable (SSVC Automatable: no). No public exploit code or confirmed active exploitation has been identified at time of analysis, and the vendor has released Firefox 157 as the fix per advisory mfsa2026-97.
Polynomial-time regular expression denial-of-service in the crmne/ruby_llm library's <think>-tag response parsing can be triggered when a model response contains many unterminated <think> tags, driving excessive CPU consumption and delaying chat-completion processing. Only deployments pinned near commit fa6f279847d6d7027814539d9c0dfc3bbdfd2a83 and running on Ruby 3.1.x are affected, since Ruby 3.2+ regex engine optimizations mitigate the backtracking; RubyLLM exposes no direct network surface of its own, so an attacker must be able to shape the upstream LLM response (malicious/compromised model endpoint, prompt injection, or interception of the model API connection). No public exploit code has been identified at time of analysis, EPSS is 0.17% (5th percentile), and there is no CISA KEV listing; the consensus signal is a low-priority issue despite the vendor-assigned 7.5 score, with real impact limited to CPU consumption and processing delay in the request handling the response.
Unauthenticated SQL injection in the third-party MediaWiki UserPageViewTracker extension allows any remote attacker who can reach a wiki running an affected version to read and exfiltrate arbitrary data from the MediaWiki database backend. The flaw is present in all versions of the extension before the vendor's fixed releases 1.46.1, 1.45.5, and 1.43.10, and per the CVSS vector (AV:N/AC:L/PR:N/UI:N) requires no authentication or user interaction; impact is confined to confidentiality (C:H) with no asserted integrity or availability effect. Publicly available exploit code exists according to the CISA SSVC assessment (Exploitation: poc), but there is no KEV listing and no confirmed active exploitation, and the EPSS score is low at 0.16% (5th percentile), consistent with a genuine but moderate-priority issue whose real-world exposure is limited by the fact that the extension is optional and absent from default MediaWiki installs.
Denial of service in the MediaWiki DataTransfer extension 1.46.0 through 1.46.x allows remote unauthenticated attackers to exhaust PHP process memory by submitting a tiny malformed .xls/OLE spreadsheet that triggers an unbounded sector-chain loop in the bundled PHPSpreadsheet library's OLERead component. The flaw is reachable through automatic spreadsheet type detection (Reader\Xls::canRead) whenever the extension processes attacker-controlled files; the CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H, 7.5 HIGH) confirms no authentication or user interaction is required and that only availability is impacted. Publicly available exploit code exists per SSVC (Exploitation: poc) even though no CISA KEV confirmed exploitation has been reported, and EPSS is low at 0.15% (percentile 4); vendor patches are available, with the extension fixed in 1.47.0 and upstream PHPSpreadsheet fixed in 5.8.1, 3.10.7, 2.4.7, and 2.1.18.
Memory exhaustion in the Apache MINA SSHD SFTP client (DefaultSftpClient, sshd-sftp component) lets an attacker-controlled SFTP server drive a client application to run out of heap and crash, because the client stores unsolicited SFTP replies that it never matches to an outstanding request and therefore never releases. The flaw affects client-side deployments of Apache MINA SSHD versions 0.9.0 through 2.19.0 and 3.0.0-M1 through 3.0.0-M5; server-only deployments are not exposed, and the victim must initiate an outbound SFTP session to a malicious or compromised server. It is rated CVSS 3.1 7.5 (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H) and is a pure denial-of-service condition of the client process, with no authentication bypass, code execution, or data disclosure; no public exploit code was identified at time of analysis and the issue is not listed as confirmed actively exploited (CISA KEV).
Out-of-bounds reads in the client-side TLS and DTLS handshake parsers of Eclipse ThreadX NetX Duo (NetX Secure) through 6.5.1.202602 allow an attacker who controls or impersonates the TLS/DTLS server to make the client parse a truncated handshake message and read a few bytes past the end of the buffer. The flaw is reachable over the network with no authentication or user interaction in the CVSS vector (AV:N/AC:L/PR:N/UI:N), but it is only triggerable when a NetX Secure client initiates a handshake toward attacker-steered server infrastructure; devices that only connect to trusted or pinned servers are substantially less exposed. No public exploit code and no CISA KEV activity were identified at time of analysis, and the impact is bounded: the independent assessment rates it confidentiality-low/availability-low (CVSS 3.1 base 6.x-equivalent) rather than the vendor's confidentiality-high score, since the over-read leaks a small amount of adjacent memory rather than yielding arbitrary disclosure or code execution.
NetX Duo (NetX Secure) through 6.5.1.202602 performs a one-byte out-of-bounds read inside the ASN.1 TLV parsing primitive that underlies every X.509 certificate operation, so a remote, unauthenticated peer can trigger it by supplying a crafted DER certificate during the TLS handshake - either against a TLS client validating a server certificate or a TLS server configured for client-certificate (mTLS) authentication. Per the independent assessment the realistic impact is limited to availability (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L), because the over-read byte is discarded and a crash requires that byte to land in unmapped or guarded memory, which is layout- and platform-dependent and frequently harmless on MMU-less embedded targets; NVD's 7.5/H score reflects the worst-case interpretation of the same vector. There is no confirmed active exploitation (CISA SSVC: Exploitation = none) and no public exploit code identified at time of analysis, though CISA rates the issue Automatable = yes and Technical Impact = partial.
Unauthenticated remote attackers can disclose sensitive information from MediaWiki installations that have the CentralAuth extension deployed at versions before 1.46.1, 1.45.5, or 1.43.10 (CVE-2026-100244, CVSS 7.5, CWE-200). Exploitation requires no authentication, no user interaction, and no special configuration beyond the extension being enabled, so the only real constraint is deployment footprint: wiki farms using CentralAuth for global/shared accounts are exposed, while standalone MediaWiki sites without the extension are not. No public exploit code has been identified at time of analysis, CISA SSVC records no observed exploitation, and vendor patches are available.
Sensitive information disclosure in the MediaWiki EventBus extension (affected build 1.47.0-alpha) allows remote, unauthenticated attackers to read event data that should be restricted to authorized event consumers, scoring CVSS 3.1 7.5 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N). Exposure is bounded by configuration and version: EventBus is not a default-bundled MediaWiki extension, so only sites that deliberately install and enable it are reachable, and only the pre-release 1.47.0-alpha branch is listed as affected, which sharply limits the realistic installed base. The independent assessment rates SSVC technical impact as partial rather than full system compromise, and no public exploit identified at time of analysis, although SSVC flags the issue as automatable and exploitation as 'none'.
Denial of service in Socket.IO Cluster Engine versions before 0.1.1 allows unauthenticated remote attackers to crash Node.js processes in clustered Socket.IO deployments by supplying crafted session IDs such as __proto__ or constructor. Only applications using @socket.io/cluster-engine are affected, and the flaw requires network reachability to the Socket.IO endpoint but no credentials or user interaction; the CVSS 3.1 vector is AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H with base 7.5. No public exploit identified at time of analysis, and the issue is fixed in 0.1.1.
Sensitive information disclosure in DetaWix Mobile Web Portal versions before v1.0.19 allows unauthenticated remote attackers to retrieve data because the application fails to properly constrain access to certain functionality (CWE-201). The CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N, base 7.5) indicates a high-confidentiality-impact flaw requiring no authentication or user interaction, making any internet-facing instance a potential target. However, no public exploit code or active exploitation has been identified at time of analysis (no KEV, no POC, no EPSS), and the product's limited install base confines realistic risk to organizations actually running this portal.
Cross-origin information disclosure in the Electron framework allows untrusted web content loaded inside an Electron application to read responses served through custom protocol schemes registered with protocol.registerFileProtocol or protocol.registerHttpProtocol when supportFetchAPI is enabled but corsEnabled is disabled. Affected are Electron builds prior to 41.10.6, 42.9.2, 43.4.1 and 44.0.0-beta.5; exploitation requires that the host application actually exposes such a scheme and loads attacker-influenced content in the same session, so severity is real but conditional rather than universal. The issue is an incomplete remediation of CVE-2026-70604, no public exploit code or active exploitation has been identified at time of analysis, and unauthenticated remote attackers can trigger it through victim interaction (CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:N/A:N, base 7.4).
Command injection in the IBM Guardium Data Protection 12.2 CLI certificate S/MIME recipient deletion routine allows an authenticated user holding a privileged CLI role to execute arbitrary shell commands with root privileges on the appliance. The flaw sits in a specific certificate/S/MIME management code path and requires high-privilege authentication plus interactive access to the command-line interface, so it is best characterized as an access-gated privilege escalation and post-compromise escalation primitive rather than a mass-exploitation threat. Patch status is "Patch available per vendor advisory" (IBM support page node/7288034); no public exploit identified at time of analysis, and no EPSS score was supplied with the intelligence feed.
Command injection in PyPA virtualenv before 21.7.12 allows an attacker who controls the environment's prompt value - via the --prompt argument, the VIRTUALENV_PROMPT environment variable, or the equivalent configuration setting - to run arbitrary cmd.exe commands on Windows with the privileges of the user who activates the generated environment. The flaw is confined to the batch activator: BatchActivator.quote() returns the prompt text unchanged, so a supplied double quote closes the `set "VAR=value"` assignment in activate.bat and any following cmd.exe operators are executed as live syntax when the victim runs the script. Coverage is limited by a local attack vector, Windows-only exposure, and mandatory victim interaction (generating and then activating the environment); no public exploit was identified at time of analysis, and a vendor fix is available in 21.7.12.
Server-side request forgery (SSRF) in the management API endpoint of HPE Networking Instant ON access points lets an authenticated remote attacker who already holds high/administrative privileges pivot from the web interface to the underlying operating system, ultimately executing arbitrary commands as a privileged user. Devices running Instant On firmware version 3.4.1.0 and earlier are affected; the flaw is network-reachable with low attack complexity and full confidentiality, integrity, and availability impact (CVSS 3.1: 7.2, AV:N/AC:L/PR:H/UI:N/S:U/C:H/I:H/A:H), but its practical reach is sharply limited by the PR:H prerequisite, meaning the attacker must first possess administrative credentials on the Instant ON management interface. No public exploit code or confirmed active exploitation was identified at time of analysis, and because exploitation is post-authentication, the realistic threat is an escalation-to-RCE step for an adversary who has already compromised or been granted AP administrator access.
Local privilege escalation in TeamViewer Full Client and Host for Windows (versions prior to 15.82) lets an authenticated low-privileged local user escalate to NT AUTHORITY\SYSTEM by winning a time-of-check-to-time-of-use race against the elevated installer's rollback routine. The attacker must already hold a low-privileged account on the host and must wait for a specific runtime event - a rollback actually occurring during a TeamViewer installation or update - after which they overwrite rollback backup files staged in a user-writable temporary directory before the elevated installer restores them; the assessed attack complexity is High (CVSS:3.1/AV:L/AC:H/PR:L/UI:R), so success is probabilistic and depends on precise timing rather than being guaranteed on a single attempt. This is a legitimate but conditional local privilege escalation: it cannot be triggered remotely, it requires a local account plus an installer/update action (UI:R), and no public exploit code or confirmed active exploitation has been identified at time of analysis. The vendor-released fix is TeamViewer Full Client and Host 15.82.
Improper authorization in WatchGuard Fireware OS's SAML login flow lets an authenticated SAML user who is scoped only to the Access Portal obtain unauthorized Mobile VPN with SSL access by submitting a specially crafted login request. The assessed vector (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:L/A:N) confirms this is a bounded privilege-escalation issue: the attacker must already hold a valid low-privilege SAML account, SAML authentication must be configured, the Access Portal must be exposed, and Mobile VPN with SSL must be provisioned (PR:L, AT:P), with no user interaction required. No public exploit code and no CISA KEV entry were identified at time of analysis.
Stored cross-site scripting in the HT Contact Form - Drag & Drop Form Builder for WordPress plugin (all versions up to and including 2.10.2) allows unauthenticated attackers to submit a crafted value through a form's Rich Text Editor field that is saved without adequate sanitization and rendered without adequate output escaping. The CVSS 3.1 vector AV:N/AC:L/PR:N/UI:N/S:C/C:L/I:L/A:N scores 7.2, reflecting that injection needs neither authentication nor interaction, but execution is deferred and only occurs when a user - most significantly a logged-in administrator reviewing form submissions, or any visitor to the affected page - loads the poisoned content. Impact is therefore real but bounded: it depends on the Rich Text Editor field being present in a deployed, publicly reachable form and on someone subsequently viewing the stored payload. No public exploit identified at time of analysis, and the plugin author has published a fix changeset referenced by the Wordfence advisory.
Deletion of arbitrary cloud snapshots on AWS, Microsoft Azure, or Google Cloud can be triggered through Barman 3.4.0/3.6.0/3.7.0 (per provider) through 3.20.0, because the snapshot deletion path reads snapshot identifiers from the backup catalog (backup.info) and passes them to the cloud provider's delete API using Barman's own credentials without verifying that those snapshots belong to the backup being removed. An authenticated, low-privileged principal who can overwrite backup.info but does not hold snapshot-delete permissions can substitute the identifiers of unrelated snapshots, causing Barman to delete any snapshot its cloud identity can reach; deletion occurs either on explicit backup deletion or automatically when a retention policy is enforced, so no further admin action is needed once the forged catalog entry exists. Exploitation is conditional, not universal: it requires a deployment in which the principal that writes the backup catalog is separate from the cloud identity Barman uses to delete snapshots, and single-identity deployments are not exploitable (CVSS 4.0 base 7.2). No public exploit code has been identified at time of analysis, and the vendor-released fix is Barman 3.20.1.
Authenticated users with any valid account on Anjvision YSSD-RTMP-H5 firmware 3.3.2.4 can silently reset the administrator password back to its default value and corrupt the device's in-memory authentication state by sending an empty-body HTTP POST to the /setUserConfig endpoint, which the web server routes through its SOAP-RPC handler without ever checking the caller's privilege level. Because the handler trusts any authenticated session as administrative, a low-privileged account is effectively sufficient to seize administrative control and to cause an availability impact on the device; only a device reload clears the corrupted state. Exploitation requires a valid account on the web interface (PR:L per the CVSS 4.0 vector and our independent CVSS 3.1 assessment of AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H), and no public exploit code or active exploitation has been identified at time of analysis - the issue was surfaced through CISA ICS advisory ICSA-26-272-05 rather than through observed attacks.
Command injection in HPE Networking Instant ON firmware versions up to and including 3.4.1.0 allows an authenticated remote attacker with high (administrator) privileges to execute arbitrary commands on the underlying operating system, resulting in full loss of confidentiality, integrity, and availability (CVSS 7.2). The flaw is network-reachable (AV:N) with low complexity (AC:L) and requires no user interaction (UI:N), but exploitation requires prior admin-level authentication (PR:H) to the management interface. No public exploit code has been identified at time of analysis, and the issue is not listed in CISA KEV.
Cleartext storage of external identity provider tokens and active session identifiers in the Devolutions Server backend database means anyone who already holds read access to those records can recover reusable credentials and replay them to impersonate users. The affected code is Devolutions Server 2026.3.5.0 and earlier; ENISA's EUVD entry narrows the affected range to all Server builds below 2026.3.7.0. Exploitation is gated by a significant prerequisite: the attacker must already have read access to the database tier, such as compromised database credentials, a leaked or stolen database backup, a privileged/malicious DBA, or a separate injection flaw, so the vendor rates it 7.2 (CVSS:3.1/AV:L/AC:H/PR:H/UI:N/S:C/C:H/I:H/A:N) while our assessment scores the same scope change with low attack complexity and limited integrity impact (AV:L/AC:L/PR:H/UI:N/S:C/C:H/I:L/A:N). No public exploit code has been identified at time of analysis, CISA SSVC records Exploitation: none with Automatable: no, and the issue is not in CISA KEV. The real risk is post-disclosure: recovered session identifiers and IdP tokens remain valid for replay, which is why the scope change drives the elevated score despite the local, high-privilege prerequisite.
Denial of service in MobilityDB 1.3.0 and earlier allows an authenticated PostgreSQL database user with the MobilityDB extension installed to crash the backend process - and thereby terminate all sessions on that instance - by supplying a crafted Well-Known Binary (WKB) payload with a negative length field. Exploitation requires only a valid low-privileged database login (CVSS AV:N/PR:L) and the ability to invoke a binary input path such as ttextFromBinary, ttextFromHexWKB, COPY ... BINARY, or binary-protocol parameters; no public exploit code has been identified at time of analysis. Vendor patches are available in MobilityDB v1.3.1 and v1.2.2.
Remote authenticated attackers can write files to arbitrary locations in Flowring Agentflow 4.0 via a path traversal flaw in the /WebAgenda/download/uploadFile.jsp endpoint, affecting versions released before March 24, 2023. The vulnerability requires a low-privilege authenticated session and no user interaction, with a CVSS 4.0 base score of 7.1. No public exploit code or active exploitation has been identified at the time of analysis.
Cross-site scripting through an InputFilter sanitizer bypass affects Joomla CMS 1.5.0 through 5.4.8 and 6.0.0 through 6.1.3, plus the standalone Joomla Framework Filter package. An authenticated user holding content-submission privileges (CVSS 3.1 PR:H) can smuggle a 'javascript:' URI past checkAttribute() by encoding characters as HTML5 entities, and the browser decodes and executes that payload when a victim later views the affected page (UI:R). Publicly available exploit code exists at proof-of-concept level (CVSS v4.0 E:P) and there is no CISA KEV entry confirming active exploitation; impact is script execution inside the victim's session context (scope changed, low confidentiality and integrity impact), making this a moderate-priority defense-in-depth filter bypass rather than a top-tier emergency.
Stored cross-site scripting (XSS) in Joomla CMS allows an authenticated user with HTML-authoring privileges - such as a content author or editor - to slip HTML data: URIs past the InputFilter::cleanAttribute routine by padding them with whitespace characters, so the payload survives sanitisation and executes in the browser of any victim who later views the stored content. Affected releases are Joomla CMS 1.5.0 through 5.4.8 and 6.0.0 through 6.1.3, along with the standalone joomla_framework_filter_package. There is no CISA KEV listing, so this is not confirmed actively exploited, and no public exploit code was identified in the supplied intelligence, although the vendor-assigned CVSS 4.0 vector does carry an exploit-maturity rating of Proof-of-Concept (E:P); the issue is not exploitable anonymously and depends on a victim rendering the injected content.
Missing authorization in the wgagent management daemon of WatchGuard Fireware OS allows an authenticated, low-privileged management account - including read-only or guest administrator roles - to crash the daemon and read files accessible to its process by sending a crafted request to the management API. The weakness sits in the daemon's session-initialization function, which fails to enforce proper privilege validation before handling attacker-controlled input; the associated path-traversal class (CWE-22) is what enables the arbitrary file read, while the crash produces a denial-of-service condition against the agent. Exploitation requires a valid management session and network reachability to the management plane (CVSS 4.0: 7.1, PR:L, VA:H, VC:L), so risk concentrates on Fireboxes with exposed or broadly accessible management interfaces, multi-admin/multi-tenant deployments, and environments where guest or read-only admin accounts are provisioned. No public exploit code has been identified at time of analysis and the issue is not listed in CISA KEV.
Crashes of PX4 Autopilot flight controllers (versions through 1.17.0) can be induced by invoking the sd_stress diagnostic system command with a -b byte count outside the documented 1-10000 range, because that value is passed unchecked into malloc() and memset() and ends up dereferencing a NULL pointer. The command shell is reachable from the adjacent MAVLink telemetry/command link (AV:A with PR:N per the CVSS vector, though practically the attacker must be able to reach that interface and run system commands), so the result is an availability-only denial of service with no code execution or data disclosure. No confirmed active exploitation (not in CISA KEV) and no public exploit code were identified at the time of analysis; the independent assessment rates this as a moderate-severity availability issue rather than a high-impact priority, and deployments that authenticate or restrict the MAVLink shell or remove sd_stress from production firmware are not exposed.
Broken object-level authorization in LimeSurvey Community Edition 7.3.0 lets an authenticated user who holds survey-creation rights tamper with questions, answers, and conditions that belong to other users' surveys. The REST survey-patching endpoint validates the caller's permission against the survey ID in the request URL (the attacker's own survey), but the underlying persistence logic resolves the target question (qid) or answer (aid) globally by ID and never confirms the object actually belongs to that authorized survey. The result is an integrity-only impact (CVSS 4.0 7.1 / assessed CVSS 3.1 AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:N) requiring a valid account plus survey-creation privileges; no public exploit code has been identified at time of analysis, and the vendor has published an upstream fix.
Authenticated users of GestSup before 3.2.62 can read private comments belonging to other users' tickets by manipulating the threadedit parameter in thread.php, because the comment loader never verifies that the requester owns the referenced ticket. Any valid account, even a low-privilege self-service one, is enough to iterate sequential comment IDs and harvest confidential ticket discussions over the network with no user interaction required. No public exploit code or active exploitation has been identified at time of analysis, and the vendor has released version 3.2.62 to close the exposure.
Denial of service of the diagnostic-tools subsystem in WatchGuard Fireware OS can be triggered by a low-privileged, authenticated user who repeatedly starts and aborts a specially crafted diagnostic task through the management web UI. The authentication requirement (PR:L per the supplied vector) and the subsystem-confined blast radius, which does not reach the firewall's core traffic-processing functions, keep this a genuine but modest-priority issue rather than a device-wide outage; no public exploit code has been identified at time of analysis and no KEV listing is present in the intelligence. Note that the vendor's CVSS 4.0 score of 7.1 records high availability impact (VA:H) on the product, while our independent assessment scores availability as low (CVSS 3.1 AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L) because the resource exhaustion is limited to the diagnostic tooling.
Authenticated low-privileged users - including read-only or guest administrator accounts - can crash the wgagent management daemon and read arbitrary files accessible to that daemon on WatchGuard Fireware OS by sending a specially crafted request to the management API's session initialization function, which is missing an authorization check. The attack is network-reachable and requires no user interaction, but it is gated behind a valid management-plane session at low privilege (PR:L), so exposure depends on how many such accounts exist and whether the management interface is reachable by the attacker. No public exploit code has been identified at time of analysis and the issue is not listed in CISA KEV; the availability impact (high) dominates the limited confidentiality loss, giving this a genuine but access-gated priority.
JIT miscompilation in Firefox's WebAssembly engine allows crafted WebAssembly delivered through attacker-controlled web content to corrupt in-process memory via a type confusion (CWE-843), producing high confidentiality impact and limited integrity impact but no availability impact. The flaw is remotely reachable at low complexity and unauthenticated (AV:N/AC:L/PR:N) but is gated by user interaction (UI:R) - the victim must load a malicious page that exercises the vulnerable JIT path - and by the content-process sandbox, so on its own it yields in-process memory disclosure rather than host compromise, which would require chaining a separate sandbox-escape bug. Mozilla has fixed it in Firefox 157, Firefox ESR 153.4, and Firefox ESR 140.17; EPSS is only 0.16% (4th percentile) and no public exploit or CISA KEV entry was identified at time of analysis.
Local privilege escalation in the npm package adm-zip through version 0.6.0: when an application extracts an archive with the non-default keepOriginalPermission=true option, the library copies the Unix permission bits recorded in each zip entry's external file attributes straight to disk via fs.chmodSync() without stripping the setuid (0o4000), setgid (0o2000) or sticky (0o1000) bits. An attacker who can supply a zip - no special access needed to craft one - can therefore plant a 04755 root-owned binary whenever the extraction runs as root (the common Docker-build, CI-runner and privileged-install posture where this flag is used), and the attacker's code executes as root once a lower-privileged user later runs that file. The issue is correctly rated Medium despite a 7.1 CVSS score because it is highly conditional (flag enabled, root extraction, setuid file preserved and executed by another user); confirmed actively exploited: no - not in CISA KEV - but a working proof-of-concept (a Python-crafted zip with external_attr 0o4755 << 16) is documented in the vendor advisory. Default usage with the flag off, or extraction as a non-root user, is not exploitable.
Denial of service in Wireshark 4.6.0 through 4.6.8 and 4.4.0 through 4.4.18 allows a malformed X11 protocol message to crash the application via a heap-based buffer overflow in the X11 dissector. Exploitation requires a user to load a crafted capture file or dissect attacker-supplied live X11 traffic, giving a local, user-interaction-dependent vector (CVSS:3.1/AV:L/AC:L/PR:N/UI:R) that, per the assessed vector, degrades only availability with no confidentiality or integrity impact - despite the vendor's 7.1 base score, which rated confidentiality high. Publicly available exploit code exists and CISA's SSVC rates the issue as proof-of-concept exploitation, not automatable, with partial technical impact; active exploitation is not indicated, and EPSS is only 0.13% (2nd percentile), so real-world risk is low. A vendor fix is available in 4.6.9 and 4.4.19.
Heap-based out-of-bounds write in the RPM Package Manager (rpm) affects Red Hat Enterprise Linux 6 through 10 and Red Hat Hardened Images, allowing a local, unauthenticated attacker to corrupt memory by crafting an RPM package that is then parsed with rpm2cpio, rpm2archive, or rpm -qlvp. The flaw arises from an integer overflow in iterReadArchiveNext() when a symlink entry declares RPMTAG_LONGFILESIZES as 0xFFFFFFFFFFFFFFFF, shrinking the buffer allocation to one byte and enabling attacker-controlled data from the cpio filesize field to overwrite heap memory. Exploitation requires user interaction to inspect the malicious package and no public exploit code has been identified at time of analysis; the CVSS 3.1 base score is 7.1.
Memory corruption in U-Boot's DHCPv6 netboot path allows an attacker on the same local network to crash or corrupt the bootloader by sending crafted DHCPv6 ADVERTISE or REPLY packets with oversized DUID (SERVERID/CLIENTID) option lengths. All U-Boot builds prior to 2026.10-rc5 that perform DHCPv6-based network boot are affected; the assessed impact is availability-focused (CVSS:3.1 AV:A/AC:L/PR:N/UI:N/C:N/I:L/A:H), with confidentiality unaffected and integrity impact limited. Exploitation requires adjacency to the booting device and only works during the brief netboot window, so devices that boot from local storage are not exposed; no public exploit and no confirmed active exploitation were identified at time of analysis.