BentoML, a Python framework for ML model serving, contains a command injection vulnerability in the docker.system_packages configuration field of bentofile.yaml files. The vulnerability affects all versions supporting this feature (confirmed in version 1.4.36) and allows attackers to execute arbitrary commands during the Docker image build process (bentoml containerize). This is a high-severity supply chain risk with a CVSS score of 7.8, requiring user interaction to trigger but achieving full command execution as root during container builds.
The Zen C compiler (versions prior to 0.4.4) crashes or enables arbitrary code execution when processing maliciously crafted .zc source files containing excessively long identifiers for structs, functions, or traits, triggering a stack-based buffer overflow (CWE-121). A proof-of-concept exploit exists per SSVC assessment, though attack complexity remains moderate as it requires local access and user interaction (CVSS:3.1/AV:L/AC:L/PR:N/UI:R). Vendor-released patch: version 0.4.4.
Ruckus Unleashed contains a remote code execution vulnerability in the web-based management interface that allows authenticated remote attackers to execute arbitrary code on the system when gateway. Rated high severity (CVSS 7.7), this vulnerability is remotely exploitable, low attack complexity. No vendor patch available.
Tandoor Recipes application versions prior to 2.6.0 expose complete database schema and access control logic through an undocumented debug parameter in the Recipe API endpoint, allowing any authenticated user to extract raw SQL queries including table structures, JOIN relationships, WHERE conditions, and multi-tenant space identifiers even in production environments with DEBUG=False. A proof-of-concept exploit is available (SSVC exploitation status: poc). The CVSS 4.0 score of 7.7 reflects network-based exploitation with no attack complexity, and SSVC indicates the vulnerability is automatable with partial technical impact.
Authenticated attackers with low-level privileges can exfiltrate sensitive database information from InvenTree open source inventory management systems prior to version 1.2.6 by abusing unvalidated filter parameters in bulk operation API endpoints. The vulnerability enables blind boolean-based data extraction through Django ORM relationship traversal, achieving high confidentiality impact with changed scope per CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:N/A:N (score 7.7). No public exploit identified at time of analysis, and vendor-released patches are available in versions 1.2.6 and 1.3.0.
SQL injection in Daylight Studio FuelCMS v1.5.2 Login.php component allows remote attackers to execute arbitrary SQL queries against the application database. The vulnerability affects the authentication mechanism, potentially enabling account enumeration, credential bypass, or unauthorized data extraction. No public exploit code or active exploitation has been confirmed at this time, though the specific attack vector suggests direct manipulation of login form parameters.
Out-of-bounds read and write in libpng's ARM/AArch64 Neon-optimized palette expansion allows remote attackers to trigger memory corruption, information disclosure, and denial of service when processing malicious PNG files. libpng versions 1.6.36 through 1.6.55 are affected on ARM platforms with Neon optimization enabled. Version 1.6.56 contains the fix. No public exploit identified at time of analysis, with SSVC framework indicating no active exploitation, non-automatable attack vector, and partial technical impact.
A stored cross-site scripting (XSS) vulnerability exists in OpenEMR's CCDA document preview functionality that allows authenticated attackers to execute arbitrary JavaScript in clinician browser sessions. OpenEMR versions prior to 8.0.0.3 are affected. The vulnerability occurs because the XSL stylesheet fails to sanitize linkHtml attributes in CCDA documents, allowing javascript: URLs and event handlers to execute when documents are previewed.
Denial of service in p11-kit's PKCS#11 RPC client allows a malicious remote token to trigger a NULL pointer dereference or undefined behavior in applications consuming derive-key operations. The flaw is reachable when an application calls C_DeriveKey against a remote token using IBM Kyber or IBM BTC derive mechanisms whose parameters are NULL, with the RPC client returning an uninitialized value. EPSS is low (0.10%) and no public exploit identified at time of analysis; SSVC marks exploitation as none but automatable.
Unauthenticated SQL injection in JS Help Desk WordPress plugin versions up to 3.0.4 allows remote attackers to extract sensitive database information via the multiformid parameter in the storeTickets() function. The vulnerability exploits improper use of esc_sql() without SQL quote encapsulation, enabling injection of additional SQL queries without requiring quote characters. CVSS scored 7.5 (High) with no public exploit identified at time of analysis and SSVC assessment indicates automatable exploitation with partial technical impact.
A security vulnerability in versions 1.2.1 (CVSS 7.5). High severity vulnerability requiring prompt remediation.
The node-forge cryptographic library for Node.js suffers from a complete Denial of Service condition when the BigInteger.modInverse() function receives zero as input, causing an infinite loop that consumes 100% CPU and blocks the event loop indefinitely. All versions of node-forge (npm package) are affected, impacting applications that process untrusted cryptographic parameters through DSA/ECDSA signature verification or custom modular arithmetic operations. CVSS 7.5 (High severity) reflects network-reachable, unauthenticated exploitation with no user interaction required. A working proof-of-concept exists demonstrating the vulnerability triggers within 5 seconds. Vendor patch is available via GitHub commit 9bb8d67b99d17e4ebb5fd7596cd699e11f25d023.
The path-to-regexp library versions 8.0.0 through 8.3.0 suffer from catastrophic regular expression denial of service via exponential regex generation when route patterns contain multiple sequential optional groups in curly-brace syntax. Remote unauthenticated attackers can trigger resource exhaustion by submitting crafted route patterns, causing application-level denial of service with CVSS 7.5 (High severity). No public exploit identified at time of analysis, and the vendor has released version 8.4.0 to address the issue.
Out-of-bounds vector access in EVerest EV charging software (everest-core versions before 2026.02.0) enables remote unauthenticated attackers to crash the charging station software or corrupt memory by sending crafted UpdateAllowedEnergyTransferModes messages from a Charging Station Management System (CSMS). CVSS 7.5 severity reflects network-accessible denial of service with high availability impact. SSVC assessment indicates no current exploitation and non-automatable attack; no public exploit identified at time of analysis.
Catastrophic backtracking in path-to-regexp versions prior to 0.1.13 enables remote denial of service attacks through specially crafted URLs containing three or more parameters within a single route segment separated by non-period characters. The vulnerability stems from insufficient backtrack protection in regex generation for routes like /:a-:b-:c, allowing unauthenticated attackers to trigger exponential computation times. SSVC framework confirms the vulnerability is automatable with partial technical impact, though no public exploit is identified at time of analysis.
CVE-2026-33870 is a security vulnerability (CVSS 7.5). High severity vulnerability requiring prompt remediation.
XML Digital Signature validation in russellhaering/goxmldsig library prior to version 1.6.0 can be bypassed due to a Go loop variable capture bug, allowing remote unauthenticated attackers to forge or manipulate XML signatures without detection. The vulnerability affects applications using Go versions before 1.22 or older go.mod configurations, enabling integrity violations in SAML authentication, document signing, and other XML-DSig implementations. EPSS score of 0.02% suggests low observed exploitation probability, with no confirmed active exploitation (not in CISA KEV). Vendor patch available in version 1.6.0.
NVMe/TCP targets are vulnerable to unauthenticated denial of service when a remote attacker sends a CONNECT command with an invalid CNTLID, triggering a kernel panic on the exposed system. The vulnerability exploits a null pointer dereference that allows any network-accessible attacker to crash the target without authentication. No patch is currently available for this high-severity flaw.
Grafana Tempo leaks S3 SSE-C encryption keys in plaintext through its /status/config endpoint, enabling unauthenticated remote attackers to retrieve encryption keys protecting trace data stored in AWS S3. The CVSS score of 7.5 reflects high confidentiality impact with network-accessible attack vector requiring no privileges or user interaction (AV:N/AC:L/PR:N/UI:N). No public exploit identified at time of analysis, though the attack path is straightforward given the information disclosure nature of the vulnerability.
The antchfx/xpath Go library prior to version 1.3.6 contains a denial-of-service vulnerability in the logicalQuery.Select function where Boolean XPath expressions that evaluate to true (such as '1=1' or 'true()') trigger an infinite loop, consuming 100% CPU resources. Remote attackers can exploit this via top-level XPath selectors without authentication, potentially disrupting any application that uses this library to process untrusted XPath queries. Upstream fix is available in commit afd4762cc342af56345a3fb4002a59281fcab494, with no public exploit code identified at time of analysis.
Drupal AI module versions 0.0.0 before 1.1.11 and 1.2.0 before 1.2.12 contain an incorrect authorization vulnerability (CWE-863) that enables resource injection attacks. The flaw allows attackers to bypass authorization controls and inject malicious resources, potentially gaining unauthorized access to AI-driven functionality or data within affected Drupal installations. No public exploit code or active exploitation has been confirmed at the time of this analysis.
DataRow.Decode in github.com/jackc/pgproto3/v2 fails to validate field length parameters, allowing a malicious or compromised PostgreSQL server to send a DataRow message with a negative field length that triggers a slice bounds out of range panic in Go applications using this library. Affected applications experience denial of service through unexpected termination when connecting to an untrusted or compromised database server. No public exploit code or active exploitation has been confirmed; however, the attack requires only network access to a PostgreSQL endpoint that the vulnerable application connects to.
The jsonparser library for Go fails to validate slice offsets when processing malformed JSON, enabling remote denial of service through crafted input that triggers negative array indices and runtime panics. The GitHub buger/jsonparser project across all versions is affected. Attackers can send specially crafted JSON payloads to applications using vulnerable versions, causing immediate service termination without requiring authentication or user interaction.
Truncated msgpack fixext format data (codes 0xd4-0xd8) decoded by shamaton/msgpack library versions across v1, v2, and v3 fail to validate input buffer boundaries, triggering out-of-bounds memory reads and runtime panics that enable denial of service. Remote attackers can craft malformed msgpack payloads to crash applications using affected library versions without requiring authentication or user interaction.
Unpublished Node Permissions module for Drupal versions prior to 1.7.0 contains an incorrect authorization vulnerability (CWE-863) that permits forceful browsing of unpublished nodes by bypassing access controls. Attackers can view content that should be restricted to specific user roles by directly accessing node URLs, circumventing the module's permission enforcement logic. No public exploit code or active exploitation has been identified at the time of analysis.
This vulnerability is a memory leak in FreeBSD's TCP stack where the tcp_respond() function fails to properly free allocated memory buffers (mbufs) when challenge ACKs are not sent in response to crafted packets. FreeBSD systems of all versions are affected. An attacker with network access (either on-path with an established connection or able to establish one, or via spoofed packets) can trigger this leak repeatedly by sending specially crafted packets that exceed rate limits, causing heap exhaustion and potential denial of service through resource depletion.
Cross-site scripting in Kestra orchestration platform versions up to 1.3.3 enables authenticated flow authors to inject arbitrary JavaScript through unsanitized Markdown rendering in flow metadata fields (description, input displayName/description). The malicious scripts execute automatically when other users view the flow in the web UI, requiring zero interaction for input.displayName fields. This vulnerability (CVSS 7.3) differs from CVE-2026-29082 and affects different components with lower interaction requirements. No public exploit identified at time of analysis, and patch availability remains unconfirmed per the advisory.
Hardcoded credentials in HCL Aftermarket DPC version 1.0.0 enable authenticated low-privilege attackers with user interaction to extract sensitive secrets from source code or insecure repositories, resulting in high confidentiality compromise and complete denial of service. CVSS score 7.3 reflects network-accessible attack requiring low privileges and user interaction. No public exploit identified at time of analysis, with SSVC framework indicating no current exploitation and non-automatable attack characteristics.
The Frontend Admin by DynamiApps plugin for WordPress contains a PHP Object Injection vulnerability affecting all versions up to and including 3.28.31. Authenticated attackers with Editor-level privileges or higher can exploit unsafe deserialization of the 'post_content' field in admin_form posts to inject malicious PHP objects and achieve remote code execution through available POP chains. This represents a critical risk for WordPress sites using this plugin with elevated user accounts.
The Blackhole for Bad Bots plugin for WordPress contains a Stored Cross-Site Scripting vulnerability that allows unauthenticated attackers to inject malicious scripts through the User-Agent HTTP header. All versions up to and including 3.8 are affected. The vulnerability stems from insufficient output escaping when displaying bot data in the admin interface, enabling arbitrary JavaScript execution when administrators view the Bad Bots log page.
Unauthenticated attackers can inject malicious scripts into Fluent Booking plugin for WordPress versions up to 2.0.01, enabling Stored Cross-Site Scripting attacks that execute in victim browsers whenever injected pages are accessed. The vulnerability stems from insufficient input sanitization across multiple parameters in LocationService.php, Booking.php, and FrontEndHandler.php. With a CVSS score of 7.2 and network-based attack vector requiring no privileges, this represents a significant threat to WordPress sites using the affected booking plugin. No public exploit identified at time of analysis, and SSVC framework indicates no current exploitation with non-automatable attack profile.
Privilege escalation in ellanetworks/core allows NetworkManager role holders to replace the production SQLite database through an unvalidated restore endpoint, escalating to Admin privileges with full access to user management, audit logs, and system configuration. The NetworkManager role was improperly granted backup/restore permissions, and the restore function accepted arbitrary SQLite files without content validation. Vendor patch available in v1.7.0 (commit 1e47682). EPSS exploitation probability is low (0.03%, 8th percentile), and no active exploitation or public POC identified at time of analysis.
Cross-user data leakage in elixir-nodejs library versions prior to 3.1.4 allows authenticated users to receive sensitive data belonging to other users through a race condition in the worker protocol's request-response handling. The lack of request-response correlation causes stale responses to be delivered to unrelated callers in high-throughput environments, potentially exposing PII, authentication tokens, or private records. No public exploit identified at time of analysis, though the vulnerability is documented in GitHub issue #100 with technical details publicly available.
Heap buffer over-read in GIMP's PCX image loader allows a remote attacker who can convince a user to open a crafted PCX file to disclose adjacent heap memory and crash the application. The flaw stems from an off-by-one error (CWE-193) and is tracked across Red Hat Enterprise Linux 6 through 9. No public exploit identified at time of analysis and EPSS is very low (0.06%, 19th percentile), with CISA SSVC rating exploitation as 'none'.
Fireshare version 1.5.1 allows authenticated remote attackers to write arbitrary files outside the intended upload directory through unsanitized path traversal in the chunked upload endpoint's checkSum parameter. The vulnerability enables attackers with valid credentials to write files to any location accessible to the Fireshare process, potentially compromising system integrity or enabling secondary attacks. No public exploit identified at time of analysis, though the vulnerability has been fixed in version 1.5.2 released by the vendor.
SQL injection in WWBN AVideo category management allows authenticated administrators to extract database contents including user credentials and private video metadata. The vulnerability resides in objects/category.php where user-supplied category title slugs are concatenated directly into SQL queries without parameterization. A working proof-of-concept demonstrates UNION-based injection to retrieve the users table. Upstream fix available via GitHub commit 994cc2b3d802b819e07e6088338e8bf4e484aae4, though no public exploit identified at time of analysis beyond the documented PoC.
SQL injection in WWBN AVideo objects/like.php allows authenticated users to read and potentially modify the entire database by injecting malicious payloads into the videos_id parameter during like/dislike actions. The vulnerability affects pkg:composer/wwbn_avideo and arises from mixing parameterized queries with direct string concatenation. A proof-of-concept UNION-based injection exists demonstrating credential extraction. Upstream fix available (PR/commit); released patched version not independently confirmed.
OpenEMR versions prior to 8.0.0.3 contain a missing authorization vulnerability in the AJAX deletion endpoint that allows any authenticated user, regardless of assigned role or privileges, to irreversibly delete critical medical data including procedure orders, answers, and specimens for any patient in the system. This is a severe integrity violation in a healthcare application handling protected health information. No evidence of active exploitation (not in CISA KEV) is currently available, though patches have been released.
Linux kernel nfnetlink_osf module fails to validate TCP option lengths in OS fingerprint definitions, allowing null pointer dereference and out-of-bounds memory reads when processing packets with malformed or missing TCP options. The vulnerability affects Linux kernel versions across multiple stable branches (6.1.x through 6.19.x and 7.0-rc5), with EPSS score of 0.02% indicating low practical exploitation probability despite the memory safety issue. No public exploit code or active exploitation has been reported.
An out-of-bounds read vulnerability in the UPnP service of TP-Link TL-WR841N v14 routers enables adjacent network attackers to crash the UPnP daemon without authentication, resulting in denial of service. Affected devices include firmware versions prior to EN_0.9.1 4.19 Build 260303 and US_0.9.1.4.19 Build 260312. Vendor patches are available. No public exploit identified at time of analysis, with CVSS:4.0 scoring 7.1 (High) reflecting adjacent network access requirements and high availability impact.
Squid prior to version 7.5 contains an out-of-bounds read vulnerability in ICP (Internet Cache Protocol) traffic handling due to improper input validation, classified as CWE-125. Remote attackers can exploit this to leak small amounts of process memory potentially containing sensitive information by sending malformed ICP requests to deployments with explicitly enabled ICP support (non-zero icp_port configuration). The vulnerability affects all versions of Squid before 7.5, and while no CVSS score or EPSS data is currently available, the information disclosure impact and remote attack vector indicate moderate to significant risk for affected deployments.
Concurrent access to an internal event queue in EVerest-core (EV charging software stack) enables remote attackers to corrupt critical data structures when CSMS GetLog or UpdateFirmware requests coincide with EVSE fault events, potentially causing information disclosure, data integrity issues, and high availability impact. The vulnerability affects all versions prior to 2026.02.0, for which a vendor patch is available. SSVC analysis indicates no current exploitation, non-automatable attack surface, and partial technical impact. EPSS data not provided; no public exploit identified at time of analysis.
The GREEN HOUSE CO., LTD. Digital Photo Frame GH-WDF10A contains active debug code that allows unauthenticated local attackers to read or write arbitrary files and execute commands with root privileges. This vulnerability affects all versions of the GH-WDF10A model and represents a critical local privilege escalation risk for any user with physical or network access to the device. While the CVSS score of 6.8 reflects medium severity due to the physical access requirement, the ability to achieve root code execution makes this a significant concern for device owners and enterprise deployments.
Improper authentication in the password-protected share handler of Kalcaddle Kodbox 1.64 allows remote attackers to bypass access controls through manipulation of the authentication function, despite high attack complexity. Public exploit code exists for this vulnerability, and no patch is currently available from the vendor.
Drupal File (Field) Paths module 7.x prior to 7.1.3 allows authenticated users to disclose other users' private files through filename-collision uploads that manipulate file URI processing, causing hook_node_insert() consumers such as email attachment modules to access incorrect file URIs and bypass access controls on sensitive files. The vulnerability affects the Drupal File (Field) Paths package as confirmed via CPE cpe:2.3:a:drupal:drupal_file_(field)_paths:*:*:*:*:*:*:*:*. No public exploit code or active exploitation data has been identified at the time of analysis.
HiJiffy Chatbot contains an authorization bypass vulnerability in the /api/v1/webchat/message endpoint that allows unauthenticated attackers to download private messages from arbitrary users by manipulating the 'visitor' parameter. The vulnerability affects all versions of HiJiffy Chatbot (as indicated by the wildcard CPE) and has been reported by INCIBE. No public exploit code or active exploitation has been confirmed at the time of analysis.
An incorrect authorization vulnerability in HiJiffy Chatbot allows unauthenticated attackers to download private messages from arbitrary users by manipulating the 'ID' parameter in the '/api/v1/download/<ID>/' endpoint. This is a classic authorization bypass enabling unauthorized access to sensitive conversation data. No public exploit code or active exploitation has been identified at the time of analysis, though the vulnerability was reported by INCIBE and affects all versions of HiJiffy Chatbot. The exposure is direct and requires only network access to the affected API endpoint.
Ruckus Access Point products contain an arbitrary file read vulnerability in the command-line interface that allows authenticated remote attackers with administrative privileges to read arbitrary. Rated medium severity (CVSS 6.9), this vulnerability is remotely exploitable, low attack complexity. No vendor patch available.
An unrestricted file upload vulnerability exists in Kalcaddle Kodbox 1.64 within the Public Share Handler component's userShare.class.php file. This allows unauthenticated remote attackers to upload arbitrary files by manipulating the Add function, potentially leading to remote code execution and system compromise. A publicly available proof-of-concept exists, and the vendor has not responded to early disclosure attempts, increasing the likelihood of active exploitation.
Mattermost Advanced Logging configuration fails to properly validate file target paths, allowing authenticated system administrators to read arbitrary files from the host system during support packet generation. The vulnerability affects Mattermost versions 11.4.0 and earlier in the 11.4.x line, 11.3.1 and earlier in the 11.3.x line, 11.2.3 and earlier in the 11.2.x line, and 10.11.11 and earlier in the 10.11.x line. An authenticated administrator with access to Advanced Logging JSON configuration can craft a malicious configuration to traverse the filesystem and extract sensitive host files through the support packet mechanism. No public exploit code has been identified at time of analysis, though exploitation requires administrative privileges and is not automatable according to CISA SSVC assessment.