Arbitrary code execution in Adobe Premiere Pro (versions 26.0.2, 25.6.4 and earlier) arises from an out-of-bounds write triggered when a victim opens a maliciously crafted project or media file, letting an attacker run code in the context of the current user. The flaw was reported by Adobe and addressed in advisory APSB26-46; it carries a CVSS 7.8 (High) score but a very low EPSS of 0.03%, and CISA SSVC records no observed exploitation. There is no public exploit identified at time of analysis, and it is not listed in CISA KEV.
Command injection in protobufjs-cli pbts tool allows arbitrary shell command execution when processing file paths with shell metacharacters. The pbts utility builds JSDoc commands by concatenating unsanitized file paths into shell strings executed via child_process.exec. Affects protobufjs-cli versions ≤1.2.0 and 2.0.0-2.0.1. Vendor-released patches available (1.2.1 and 2.0.2). CVSS 7.8 (High) but requires local access with user interaction (AV:L/UI:R), limiting remote exploitation. No EPSS data or KEV listing indicates this is not yet widely exploited despite public disclosure and available fixes.
Arbitrary code execution in Adobe After Effects 26.0, 25.6.4 and all earlier versions arises from a stack-based buffer overflow (CWE-121) triggered when a victim opens a maliciously crafted project or media file, running code in the context of the current user. Reported by Adobe and tracked in APSB26-48, it carries a CVSS 7.8 (High) with a local attack vector and required user interaction; there is no public exploit identified at time of analysis, and the EPSS score is 0.01% (1st percentile). SSVC indicates no observed exploitation but total technical impact, making this a patch-on-schedule priority rather than an emergency.
Local privilege escalation in Ivanti Endpoint Manager agent allows authenticated users to gain SYSTEM-level privileges via incorrect file or registry permissions. Affects all versions prior to 2024 SU6. Vendor has released a patch (version 2024 SU6). No evidence of active exploitation or public POC identified at time of analysis, though EPSS data not available. Organizations running EPM agents on managed endpoints should prioritize patching given the high CVSS score (7.8) and potential for lateral movement across enterprise environments.
Race condition in Ivanti Secure Access Client enables local privilege escalation to SYSTEM from low-privileged accounts. Affects versions before 22.8R6. An authenticated local user can exploit timing vulnerabilities in the client software to gain complete system control. While limited to local attack vector (requires existing access to the target system), the low attack complexity (AC:L) and lack of user interaction requirement (UI:N) make this exploitable once local access is achieved. No public exploit code identified at time of analysis, and EPSS risk scoring not yet available for this 2026 CVE.
Remote code execution in protobufjs (npm package) versions ≤7.5.5 and 8.0.0-8.0.1 allows attackers to inject and execute arbitrary JavaScript by supplying a malicious protobuf schema with crafted default values in bytes fields. When applications load untrusted protobuf descriptors and call toObject() with defaults enabled, attacker-controlled expressions are emitted into generated conversion functions and executed in the application context. Vendor-released patches are available in versions 7.5.6 and 8.0.2. No public exploit code identified at time of analysis, though the vulnerability is straightforward to weaponize given the clear preconditions in the advisory.
Server-Side Request Forgery in Xibo CMS versions prior to 4.4.1 lets an authenticated user with Library upload permissions coerce the CMS server into issuing arbitrary HTTP requests to internal or external hosts. This enables internal network scanning, reaching cloud instance-metadata endpoints such as AWS IMDS, pivoting to unauthenticated internal services, and data exfiltration. No public exploit identified at time of analysis; EPSS is low (0.03%, 9th percentile) and CISA SSVC records exploitation as none.
Blind SQL injection in WP Travel plugin versions ≤11.4.0 allows authenticated attackers with low-level privileges to extract sensitive database contents through time-based or boolean queries. The vulnerability enables cross-scope confidentiality breach with high impact (CVSS:C:H), permitting unauthorized access to all WordPress database information including user credentials, private travel booking details, and payment data. EPSS data unavailable; no CISA KEV listing indicates exploitation remains targeted rather than widespread. Patchstack's inclusion in their vulnerability database suggests active researcher interest and potential proof-of-concept development.
Authorization bypass in Heym's workflow execution engine (versions before 0.0.21) allows any authenticated user to run workflows belonging to other users by referencing the victim's workflow UUIDs, exposing their outputs and triggering side-effecting nodes. The root cause is missing access validation on subworkflow and execute-node references (CWE-863), fixed upstream in pull request #93. No public exploit has been identified at time of analysis, and EPSS scores it at just 0.05% with CISA SSVC marking exploitation as 'none' - but SSVC rates the technical impact as total.
Blind SQL injection in BEAR woo-bulk-editor plugin for WordPress up to version 1.1.7.1 allows high-privilege authenticated administrators to extract database contents through specially crafted SQL queries. The scope change in CVSS (S:C) indicates potential impact beyond the plugin itself, enabling access to other WordPress data or resources. No public exploit code or active exploitation indicators identified at time of analysis, but Patchstack public disclosure increases weaponization risk.
Session fixation in Pandora FMS versions 777-800 enables session hijacking when attackers supply crafted session IDs to users. Successful exploitation grants attackers complete access to victim user sessions with high confidentiality and integrity impact. No public exploit code identified at time of analysis, though attack complexity is low with network-based delivery requiring only user interaction (CVSS 7.6).
SQL injection in Pandora FMS versions 777-800 allows authenticated attackers with low privileges to exfiltrate or manipulate database contents via the graph container parameter. Attack complexity is high with present attack techniques, requiring specific timing conditions. No active exploitation confirmed per CISA KEV, and EPSS data not provided. Vendor advisory available from PandoraFMS confirms the vulnerability affecting a narrow version range spanning approximately builds 777 through 800.
Remote unauthenticated attackers can crash Zyxel NWA1100-N access points running customized firmware version 1.00(AACE.1)C0 by sending malformed HTTP requests that trigger buffer overflows in five distinct web server functions (formWep, formWlAc, formPasswordSetup, formUpgradeCert, formDelcert). The vulnerability enables denial-of-service attacks with high CVSS 7.5 severity but is limited to an end-of-life product according to Zyxel's reference documentation. No public exploit code identified at time of analysis, and EPSS data is unavailable for this recent CVE.
Denial of service in the Microsoft Windows TCP/IP stack (Windows 11 24H2/25H2/26H1 and Windows Server 2025) allows a remote, unauthenticated attacker to crash the affected system by triggering a null pointer dereference over the network. No public exploit has been identified at time of analysis and the EPSS probability is low (0.14%, 34th percentile), but the flaw is network-reachable with no privileges or user interaction and CISA SSVC rates it automatable. Impact is limited to availability - no code execution or data disclosure.
Remote code execution in HPE Aruba Networking Wireless Operating System (AOS-8 and AOS-10) allows an unauthenticated remote attacker to overflow a heap buffer in a network management service and run arbitrary code as a privileged OS user, or trigger a denial-of-service crash of the affected process. It affects Aruba mobility controllers and gateways across multiple 8.x and 10.x branches. No public exploit identified at time of analysis and it is not in CISA KEV; EPSS is low (0.12%, 30th percentile), and notably the vendor CVSS vector scores only availability impact despite the RCE claim.
Information disclosure in Microsoft Power Automate for Desktop (all versions from 1.0.0.0 up to but not including 2.67) enables an authenticated, low-privileged attacker operating over the network to access sensitive data they are not authorized to view. The flaw, rooted in CWE-200 improper exposure of sensitive information, carries a CVSS 6.5 score reflecting high confidentiality impact with no integrity or availability consequence. No active exploitation has been confirmed - EPSS stands at 0.11% and CISA KEV listing is absent - but patch availability from Microsoft makes remediation straightforward.
SQL injection in the AIWU AI Chatbot WordPress plugin (versions ≤1.4.17) allows remote unauthenticated attackers to extract sensitive database contents via the getListForTbl() function due to unsanitized user input in SQL queries. Partial mitigation exists in version 1.4.11+ through administrator-only nonce protection, but the underlying SQL injection vulnerability persists. CVSS 7.5 (High) reflects network-accessible unauthenticated exploitation with high confidentiality impact. Wordfence provides detailed vulnerable code references across multiple plugin files including controller.php, req.php, and model.php. No evidence of active exploitation (not in CISA KEV) at time of analysis.
Improper authorization in Adobe Commerce (Magento) 2.4.9-beta1, 2.4.8-p4, 2.4.7-p9, 2.4.6-p14, 2.4.5-p16, 2.4.4-p17 and earlier lets a remote unauthenticated attacker (CVSS PR:N) bypass a security control and gain unauthorized write access to the affected system, with no user interaction. There is no public exploit identified at time of analysis and the EPSS score is low (0.09%), but the network-reachable, no-auth, no-interaction vector combined with Adobe's own reporting warrants prompt patching. This is not in CISA KEV and shows no confirmed exploitation.
Improper authorization in Adobe Commerce (Magento) lets remote unauthenticated attackers bypass security controls and gain unauthorized write access across builds 2.4.9-beta1, 2.4.8-p4, 2.4.7-p9, 2.4.6-p14, 2.4.5-p16, 2.4.4-p17 and earlier. The flaw (CWE-863) requires no user interaction and is exploitable over the network per its CVSS:3.1 vector, though there is no public exploit identified at time of analysis and it is not on CISA KEV. EPSS is low (0.09%, 26th percentile), but CISA SSVC marks it 'Automatable: yes', meaning any working exploit could be scaled easily.
Denial-of-service in HPE Aruba Networking Wireless Operating System (AOS-8) lets an unauthenticated remote attacker crash a network management service by sending specially crafted packets, terminating the service process and disrupting normal controller operations. Affected releases include AOS 8.10.x, 8.12.x, and 8.13.x up to the fixed builds. There is no public exploit identified at time of analysis, and EPSS is low (0.08%), consistent with SSVC exploitation status of 'none' - this is a patch-worthy availability issue rather than an active threat.
Denial of service in the Windows Internet Key Exchange (IKE) protocol implementation allows remote unauthenticated attackers to exhaust host memory across all supported Windows client and server releases (Windows 10 1607 through Windows 11 26H1, Windows Server 2012 through Server 2025). The flaw is a memory leak (CWE-401) triggered over the network with no privileges or user interaction, but impact is limited to availability with no data exposure. There is no public exploit identified at time of analysis, and EPSS exploitation probability is very low (0.08%, 23rd percentile).
Denial-of-service in Adobe Commerce (Magento) versions 2.4.9-beta1, 2.4.8-p4, 2.4.7-p9, 2.4.6-p14, 2.4.5-p16, 2.4.4-p17 and earlier lets a remote unauthenticated attacker crash the application by exercising a bundled vulnerable third-party component, taking the storefront offline. The flaw carries CVSS 7.5 with an availability-only impact and no confidentiality or integrity effect; there is no public exploit identified at time of analysis, and EPSS is low at 0.07%. It was reported by Adobe and addressed in security bulletin APSB26-49.
Remote code execution in the Windows Native WiFi Miniport Driver allows an unauthorized attacker on an adjacent network to win a timing race and run arbitrary code on affected Windows 10, Windows 11, and Windows Server systems. The flaw stems from improper synchronization of a shared resource in the kernel-mode wireless driver and carries a total (C:H/I:H/A:H) technical impact. No public exploit has been identified and EPSS is very low (0.07%), but successful exploitation grants full kernel-level compromise without authentication.
SQL injection in the Court Reservation - Manage Your Court Bookings Online WordPress plugin (versions ≤ 1.10.11) lets unauthenticated attackers inject arbitrary SQL through the 'id' parameter, enabling extraction of sensitive database contents such as user credentials and booking data. The flaw stems from an unescaped, unprepared query and is network-reachable with no authentication or user interaction. No public exploit identified at time of analysis, and EPSS probability is low (0.06%), but SSVC flags exploitation as automatable with partial technical impact.
Information disclosure in the Windows TCP/IP networking stack allows a remote, unauthenticated attacker to read sensitive memory contents over the network by triggering a use-after-free condition. All supported Windows client and server releases (Windows 10 1607 through Windows 11 26H1, and Windows Server 2012 through 2025) are affected. Microsoft-reported with no public exploit identified at time of analysis; EPSS is low (0.06%, 19th percentile) and CISA SSVC records exploitation status as none.
Unauthenticated remote attackers can modify or delete arbitrary memory records in mem0 1.0.0 server by directly calling unprotected API endpoints including PUT /memories/{memory_id}. The vulnerability stems from complete absence of authentication and authorization controls on critical memory management functions, allowing data manipulation and loss without any verification of requester identity. EPSS score of 0.06% (18th percentile) indicates low exploitation probability in the wild, and no public exploit code or active exploitation (CISA KEV) has been identified at time of analysis.
Local privilege escalation in Azure Monitor Agent Metrics Extension (versions below 1.42.0) allows an authenticated low-privileged local attacker to elevate privileges by exploiting an untrusted search path condition (CWE-426). The agent process resolves executable or library paths from an attacker-controllable location, enabling substitution of malicious code that executes in a higher-privileged security context (S:C). No public exploit code has been identified at time of analysis, and EPSS stands at 0.05% (17th percentile), indicating minimal observed exploitation pressure.
Remote denial of service in protobuf.js (npm package) allows unauthenticated attackers to crash Node.js processes by sending crafted protobuf payloads with deeply nested structures. The vulnerability affects the binary decoding path where unbounded recursion exhausts the JavaScript call stack. Version 7.5.6 and 8.0.2 patches are available. Applications decoding untrusted protobuf data over network APIs, message queues, or file uploads are at immediate risk. CVSS 7.5 (High) reflects network attack vector with no authentication required.
Arbitrary file read in the esm.sh CDN build server (Go implementation, versions ≤ 137) lets a remote attacker exfiltrate sensitive host files by publishing a malicious npm package. The esbuild plugin re-maps module paths using the package.json 'browser' field without re-validating the sandbox, so '../' sequences escape the package directory and file contents are returned inside the bundled JS and source-map sourcesContent. A working PoC exists (published npm package chess-sec-utils1) reading /etc/hostname, /etc/os-release and /etc/environment, but there is no public exploit identified as actively used and EPSS risk is low (0.04%).
Digest authentication bypass in Espressif's arduino-esp32 WebServer (all versions prior to 3.3.8) lets an attacker who holds any single valid digest response replay it to reach a different protected endpoint, defeating per-resource access control. The flaw is CWE-287 (improper authentication) because the server trusts the URI value inside the client's Authorization header instead of binding the hash to the actually-requested resource. There is no public exploit identified at time of analysis, EPSS is very low (0.04%, 13th percentile), and CISA SSVC marks exploitation as none but automatable.
Denial of service in multiparty (Node.js multipart/form-data parser) versions ≤4.2.3 crashes Node.js processes when attackers send crafted form uploads with field names matching JavaScript Object prototype properties (__proto__, constructor, toString). CVSS 7.5 (High) with network vector and no authentication required. No public exploit code identified at time of analysis, but exploitation is trivial given the straightforward prototype pollution attack pattern. Services accepting file uploads via multiparty are immediately affected until upgraded to 4.3.0+.
Remote denial of service in Deskflow servers (versions prior to 1.26.0.167) running with TLS enabled - the default - lets any unauthenticated TCP peer stall input delivery to all connected screens. Because SecureSocket::secureAccept calls a blocking Arch::sleep(1) on the single multiplexer worker thread whenever a connection's first bytes fail to parse as a valid TLS ClientHello, one malformed handshake freezes mouse, keyboard, and clipboard traffic for ~1 second, and a sustained drip of ≥1 bad connection per second renders the server unusable. EPSS is low (0.04%, 12th percentile) and there is no public exploit identified at time of analysis, though the fix commit is public and the trigger is trivial.
Uncontrolled resource consumption in Adobe Commerce (Magento) versions 2.4.9-beta1, 2.4.8-p4, 2.4.7-p9, 2.4.6-p14, 2.4.5-p16, 2.4.4-p17 and earlier lets a remote attacker exhaust system resources and force an application-level denial-of-service without any authentication or user interaction. The flaw carries a CVSS 7.5 (availability-only) rating and, per Adobe's APSB26-49 advisory, affects the e-commerce platform's server-side request handling. There is no public exploit identified at time of analysis and the EPSS score is very low (0.04%, 12th percentile), indicating no observed exploitation activity to date.
Denial of service crashes multiparty Node.js parser versions ≤4.2.3 when processing malformed percent-encoded filename* parameters in multipart/form-data uploads. Attackers can remotely crash any Node.js service using vulnerable multiparty versions by sending a single crafted HTTP request with no authentication required (CVSS:3.1 AV:N/AC:L/PR:N/UI:N). No public exploit identified at time of analysis, but exploitation is trivial given the straightforward attack vector. Vendor-released patch: multiparty@4.3.0.
Denial-of-service in Adobe Commerce (Magento) 2.4.4-p17, 2.4.5-p16, 2.4.6-p14, 2.4.7-p9, 2.4.8-p4, 2.4.9-beta1 and earlier allows a remote attacker to exhaust system resources and render the application unavailable without any authentication or user interaction. The flaw is an Uncontrolled Resource Consumption (CWE-400) issue reported by Adobe (APSB26-49); it carries a CVSS 7.5 (availability-only impact) with no confidentiality or integrity effect. There is no public exploit identified at time of analysis and the EPSS score is very low (0.04%), indicating no observed weaponization to date.
Regular expression denial of service in multiparty (npm package) versions 4.2.3 and below allows remote unauthenticated attackers to block the Node.js event loop for seconds via crafted Content-Disposition headers in multipart uploads. The vulnerability triggers catastrophic backtracking in the filename parameter parser with headers as small as 8 KB. Fixed in multiparty 4.3.0. EPSS data not available; no active exploitation confirmed at time of analysis, but the low complexity (AV:N/AC:L/PR:N/UI:N) and availability of detailed public advisory increase weaponization risk for any web service accepting file uploads through this library.
Complete process crash in Dalfox REST server v2.12.0 and earlier allows remote unauthenticated attackers to terminate the entire scan engine via a single HTTP request. The flaw stems from a closed-channel write panic in pkg/scanning/parameterAnalysis.go where two sequential worker stages share a single results channel, causing a Go runtime panic when the second stage attempts to write to the already-closed channel. Default deployments run without authentication (no --api-key), making the vulnerability remotely exploitable. CVSS 7.5 (High). Patched in v2.13.0 per GitHub advisory GHSA-2g4x-fq3j-cgq4. No CISA KEV listing or public exploit code identified at time of analysis.
Remote unauthenticated attackers can access confidential data from other users' chat sessions in Spring AI applications due to insecure default configuration in the chat memory component. The vulnerability allows network-based exploitation with no authentication required (CVSS:3.1 AV:N/AC:L/PR:N/UI:N) and impacts confidentiality only (C:H/I:N/A:N), enabling cross-user data leakage in multi-tenant AI chat implementations. Reported by VMware, affecting Java-based Spring AI deployments where developers have not explicitly configured chat memory isolation.
Sensitive user data disclosure in Apple iOS and iPadOS before 18.7.3 and before 26.2 stems from an inconsistent user-interface state that allows a malicious or crafted app to access data it should not. The flaw, reported by Apple's own product security team and fixed across the iOS 18.7.x and 26.x branches, carries a CVSS of 7.5 (high confidentiality impact) but an EPSS of just 0.03%, and there is no public exploit identified at time of analysis. CISA SSVC rates exploitation as none, indicating a real but low-urgency issue rather than an actively targeted one.
Denial-of-service in HPE Aruba Networking Wireless Operating System (AOS-8 and AOS-10) allows a remote unauthenticated attacker to crash a critical system process by sending specially crafted network messages to a protocol-handling service (CWE-20, insufficient input validation). Reported by HPE with no public exploit identified at time of analysis; EPSS is very low (0.03%, 9th percentile) and CISA SSVC records exploitation status as none, though the flaw is rated automatable. Impact is limited to availability (CVSS 7.5, A:H) with no confidentiality or integrity loss.
Denial-of-service in HPE Aruba Networking Wireless Operating System (AOS-8 and AOS-10) allows a remote unauthenticated attacker to crash a critical system process by sending specially crafted network messages to a protocol-handling service. Because input validation is insufficient, a successful attack terminates the process and disrupts wireless network operations without affecting confidentiality or integrity. There is no public exploit identified at time of analysis, and the EPSS score is very low (0.03%), indicating no observed exploitation activity.
Information disclosure in Apache Tomcat's LockOutRealm component allows remote attackers to bypass account-lockout tracking because usernames are compared with inconsistent case sensitivity (CWE-178), enabling authentication-related information leakage across Tomcat 7.0.0-7.0.109, 8.5.0-8.5.100, 9.0.0.M1-9.0.117, 10.1.0-M1-10.1.54 and 11.0.0-M1-11.0.21. The CVSS 3.1 vector (7.5, C:H only) scopes the impact to confidentiality with no integrity or availability effect, consistent with the 'Information Disclosure' tagging. There is no public exploit identified at time of analysis, EPSS is negligible (0.02%, 5th percentile), and CISA SSVC records exploitation as none.
Denial of service in Apache Tomcat 9.x, 10.1.x, and 11.0.x allows remote unauthenticated attackers to exhaust server resources due to missing limits or throttling on a resource allocation path (CWE-770). Affected versions span 11.0.0-M1 through 11.0.21, 10.1.0-M1 through 10.1.54, and 9.0.0.M1 through 9.0.117, with older unsupported branches also implicated per the EUVD entry. No public exploit identified at time of analysis, and the EPSS score is very low (0.02%, 5th percentile), but the SSVC framework flags the issue as automatable with partial technical impact.
SQL injection in the Elixir postgrex library allows local attackers with control over PostgreSQL LISTEN/UNLISTEN channel names to execute arbitrary SQL commands including DDL and DML operations. The Postgrex.Notifications module (versions 0.16.0 through 0.22.1) fails to escape double-quote characters in channel arguments, enabling attackers to break out of quoted identifiers and chain multi-statement payloads such as DROP TABLE commands. Vendor patch available in version 0.22.2 per GitHub advisory GHSA-r73h-97w8-m54h. No public exploit code or CISA KEV listing identified at time of analysis, though the technical details and patch diff are publicly disclosed.
Session persistence after credential change in SillyTavern before 1.18.0 lets an attacker holding a stolen session cookie retain full account access even after the victim resets their password. Because sessions are stateless and stored entirely in a signed client-side cookie with a default 400-day lifespan, the change-password and account-recovery endpoints update the password hash but never revoke outstanding tokens, defeating the standard remediation for cookie theft. No active exploitation is recorded (not in CISA KEV) and EPSS is very low at 0.02%, but public proof-of-concept reproduction steps are documented in the GHSA advisory.
Application denial-of-service in Adobe Commerce (Magento) affects versions 2.4.9-beta1, 2.4.8-p4, 2.4.7-p9, 2.4.6-p14, 2.4.5-p16, 2.4.4-p17 and earlier, where an unauthenticated remote attacker can trigger uncontrolled resource consumption (CWE-400) to exhaust system resources without any user interaction. There is no public exploit identified at time of analysis and the EPSS probability is very low (0.01%, 2nd percentile), so while the CVSS availability impact is High (7.5), real-world weaponization appears unlikely in the near term. Impact is limited to availability, with no confidentiality or integrity effect.
Application denial-of-service in Adobe Commerce (Magento) affects versions 2.4.9-beta1, 2.4.8-p4, 2.4.7-p9, 2.4.6-p14, 2.4.5-p16, 2.4.4-p17 and earlier, where an Uncontrolled Resource Consumption flaw (CWE-400) lets a remote, unauthenticated attacker exhaust server resources without any user interaction. The CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N/A:H) reflects a high availability impact with no confidentiality or integrity loss. There is no public exploit identified at time of analysis, and the EPSS score is very low (0.01%, 2nd percentile), indicating minimal observed exploitation interest despite the network-reachable attack surface.
Uncontrolled resource consumption in Adobe's CAI (Content Authenticity Initiative) Content Credentials SDK versions 0.78.2, 0.7.0 and earlier lets a remote unauthenticated attacker exhaust system resources and crash the application (availability-only denial-of-service). No public exploit is identified at time of analysis and EPSS probability is negligible (0.01%), but SSVC flags the attack as automatable, so exploitation could be scripted once a vector is understood. Confidentiality and integrity are not impacted.
Unauthenticated arbitrary file read in dalfox REST API server mode allows remote attackers to exfiltrate sensitive files from the host filesystem. The vulnerability chains two flaws: missing authentication middleware when no API key is set (default configuration), and unsanitized deserialization of the `custom-payload-file` JSON parameter directly into the scan engine. Remote attackers can supply any file path (e.g., `/etc/passwd`, `~/.ssh/id_rsa`, cloud credential files) and the engine reads each line, embeds it as an XSS payload, and transmits it to an attacker-controlled HTTP endpoint via dalfox's own scan traffic. No exploit code is publicly identified at time of analysis; vendor-released patch available in version 2.13.0.
Prototype pollution in protobufjs allows denial of service through corrupted JavaScript built-ins when parsing untrusted schemas. Attackers who control protobuf schemas or JSON descriptors can write to inherited object properties on global constructors, causing process-wide state corruption that persists until restart. CVSS 7.5 (High) with network vector and no authentication required, but real-world risk is limited to applications parsing schemas from untrusted sources-applications only decoding untrusted message payloads with trusted schemas are not affected. Vendor-released patches available: v7.5.6 and v8.0.2. No active exploitation confirmed (not in CISA KEV), and no public exploit code identified at time of analysis.