Overview of incoming advisories.
2186 results found
CVE-2026-81719 affects openssl_encrypt versions prior to 1.4.9, where third-party plugins were executed with insufficient security controls. The plugin signature policy defaulted to WARN instead of ENFORCE, allowing unsigned or unverifiable non-built-in plugins to be compiled and executed in the host process at import time, before the runtime sandbox was installed. The only default protection was an incomplete and bypassable AST denylist. If a user is induced to load a malicious plugin, an attacker can achieve arbitrary code execution with the privileges of the user running openssl_encrypt. The vulnerability has been fixed in version 1.4.9 by changing the default signature policy to ENFORCE for non-built-in plugins, preventing unsigned plugins from executing without explicit user approval.
View original advisory →A server-side request forgery (SSRF) vulnerability has been identified in ddfourtwo sentry-selfhosted-mcp version 0.4.0. The flaw exists in an unknown function of the raw_sentry_api component, where manipulation of the endpoint argument allows SSRF attacks. The vulnerability can be exploited remotely without requiring physical access. A public exploit has already been released, increasing the risk of active exploitation. The project maintainer was notified via an issue report but has not yet responded or issued a patch. The lack of vendor response and public exploit availability makes this a significant risk for users of the affected software.
View original advisory →An Improper Authorization vulnerability has been identified in 3DPassport, a component of 3DSwymer by Dassault Systèmes. The vulnerability affects releases from 3DEXPERIENCE R2023x through R2026x. If exploited, an attacker could gain unauthorized access to certain user accounts. The flaw is classified as an Improper Authorization issue, meaning access controls are insufficiently enforced. This could allow privilege escalation or account takeover for some users. The vulnerability is tracked as CVE-2026-16279 and has been assigned a high criticality rating. Dassault Systèmes has published a security advisory through their Trust Center. Users running affected versions are advised to review the advisory and apply any available patches or mitigations promptly.
View original advisory →CVE-2026-32479 describes an unauthenticated SQL injection vulnerability affecting the WordPress plugin Visitor Traffic Real Time Statistics Pro in versions 11.17 and below. The vulnerability allows unauthenticated attackers to inject malicious SQL queries, potentially leading to unauthorized access to the underlying database. No authentication is required to exploit this flaw, making it particularly dangerous. The issue was reported via NVD and corroborated by Patchstack's vulnerability database. Users of the affected plugin are advised to update to a patched version immediately. The vulnerability carries a high criticality rating given the unauthenticated nature of the exploit vector.
View original advisory →CVE-2026-81091 is a Server-Side Request Forgery (SSRF) vulnerability in the proxy middleware of mcp-use's inspector component. The vulnerable function mountMcpProxy in mcp-proxy.ts reads the target URL from the X-Target-URL header or __mcp_target parameter and forwards requests without validating the destination host. This allowed callers to direct the server to issue requests to loopback, link-local, and private network addresses, as well as hostnames that resolve to them. Additionally, redirect responses from the destination were not re-validated, enabling redirect-based bypass. An attacker could exploit this to access internal services and read responses from addresses only reachable from the host machine. The fix introduces an isSafeProxyTarget function that checks resolved addresses against private, loopback, and link-local ranges and limits redirect following.
View original advisory →A vulnerability exists in the MongoDB C++ Driver related to improper handling of caller-supplied namespace identifiers. The flaw allows special characters to be embedded in namespace identifiers without adequate validation. Applications that construct namespace identifiers from untrusted input are susceptible to having their operations redirected to unintended targets. This can lead to limited unauthorized read and write access to data belonging to other logical tenants of the affected application. The issue is a form of namespace injection or tenant isolation bypass. The vulnerability is tracked as CVE-2026-81522 and has been addressed in the MongoDB C++ Driver release r4.5.1. Applications using the driver with multi-tenant architectures are at heightened risk if they do not validate user-supplied input before constructing namespace identifiers.
View original advisory →NLTK versions through 3.10.3 contain a path traversal vulnerability in model-artifact APIs that bypass pathsec enforcement by using raw file operations on caller-controlled paths. The vulnerability affects TransitionParser, AveragedPerceptron, PerceptronTagger, and maxent parameter APIs. Attackers can exploit this flaw to read or write files outside allowed sandbox roots even when pathsec is enabled. The issue stems from the use of raw file operations on caller-controlled paths rather than properly sanitized paths. This represents a significant security risk for applications that rely on NLTK's pathsec mechanism for sandbox enforcement. The vulnerability has been assigned CVE-2026-81726 and is documented in both the NVD and GitHub Security Advisories. Organizations using NLTK for natural language processing tasks should prioritize patching or mitigating this vulnerability.
View original advisory →CVE-2026-81699 affects openssl_encrypt versions before 1.4.9, which fail to properly validate key derivation function (KDF) costs in crafted files. Attackers can supply malicious files containing excessive KDF parameters to trigger unbounded memory and CPU exhaustion. The vulnerability is exploitable during pre-authentication processing, meaning no valid credentials are required to launch the attack. Successful exploitation can cause the affected process to crash or become unresponsive. This represents a Denial of Service (DoS) risk that can be triggered remotely before any password verification occurs. The fix is available in openssl_encrypt version 1.4.9 and later.
View original advisory →OpenRemote versions before 1.28.0 contain a cross-realm information disclosure vulnerability in the Notification REST API. The flaw allows per-realm tenant administrators with read:admin credentials to access sent notifications from all tenants, not just their own realm. Exploitation requires only a zero-parameter GET request to the notification endpoint, making it trivially easy to abuse. Sensitive notification metadata and message content from all realms can be retrieved by an attacker with access to any single realm. The vulnerability has been addressed in OpenRemote version 1.28.0. This represents a significant access control failure in a multi-tenant architecture. The issue is documented in both the NVD and GitHub Security Advisories. Organizations running OpenRemote in multi-tenant deployments are particularly at risk. The vulnerability could expose confidential communications between administrators and users across all tenants.
View original advisory →A critical unauthenticated PHP Object Injection vulnerability has been identified in the Geo Controller WordPress plugin affecting versions 8.9.8 and below. The vulnerability allows unauthenticated attackers to inject PHP objects, potentially leading to remote code execution or other severe impacts depending on available POP chains in the environment. The flaw is documented under CVE-2026-78286 and has been reported via both the NVD and Patchstack vulnerability databases. No authentication is required to exploit this vulnerability, significantly increasing its risk exposure. WordPress site administrators using the Geo Controller plugin (cf-geoplugin) should update to a patched version immediately. The vulnerability was flagged with a high criticality rating. Patchstack has published additional details and patch information for affected users.
View original advisory →A critical unauthenticated SQL injection vulnerability has been identified in the Epayco WordPress plugin versions 8.4.6 and below. The flaw allows unauthenticated attackers to interact directly with the database, potentially exposing sensitive data or enabling full database compromise. No authentication is required to exploit this vulnerability, making it particularly dangerous in production environments. The vulnerability is tracked under CVE-2026-78260 and has been documented by both the NVD and Patchstack. Users of the Epayco payment gateway plugin for WordPress are advised to update to a patched version immediately. SQL injection vulnerabilities of this nature can lead to data exfiltration, authentication bypass, and in some cases remote code execution depending on server configuration.
View original advisory →CVE-2026-81335 affects Baserow versions prior to 2.3.1, exposing a critical authorization bypass in the Application Builder's data source dispatch mechanism. The dispatch and record-name API views are configured with a permissive class that allows any caller, including unauthenticated requests, to reach the handler. The DataSourceService.dispatch_data_sources method calls check_multiple_permissions but never evaluates or stores the denial mapping it returns, allowing execution to continue regardless of permission outcome. As a result, the data source is dispatched using the integration's own credentials, leaking rows and fields to unauthenticated callers. Data source identifiers are small integers, making them trivially enumerable and increasing the risk of mass data exposure. The fix in version 2.3.1 passes the raise_exception flag to the permission check, ensuring denials halt execution. This vulnerability enables unauthenticated data exfiltration from any Baserow Application Builder data source. Organizations using Baserow should upgrade to version 2.3.1 immediately.
View original advisory →CVE-2026-77438 affects Trilium, an open-source hierarchical note-taking application, in versions up to and including 0.103.0. The public share-search endpoint fails to enforce per-note shareCredentials and shareHiddenFromTree access controls. An unauthenticated attacker can exploit this flaw to read titles, tree paths, and full content of password-protected shared notes. The endpoint only authorizes the ancestor note in the request and then performs a full-text search across the entire published subtree without re-checking individual note permissions. This allows an attacker to use the endpoint as a boolean oracle to confirm arbitrary substrings and reconstruct protected note contents. The vulnerability enables enumeration of hidden notes and bypass of share password protections. No authentication is required to exploit this issue. The vulnerability has been fixed in Trilium version 0.104.0.
View original advisory →CVE-2026-81093 describes a Server-Side Request Forgery (SSRF) vulnerability in the Apify MCP Server's get-html-skeleton tool. The tool accepted caller-supplied URLs and only validated their syntax (http/https scheme and parseable format) without inspecting the hostname or resolved IP address. This allowed loopback, link-local, and private IP ranges to pass validation, including cloud provider instance metadata endpoints (e.g., 169.254.169.254). Any caller of the MCP server could exploit this to make the server fetch internal-only endpoints and return the response, potentially exposing sensitive instance credentials and internal services. The vulnerability resided in src/tools/common/get_html_skeleton.ts using the isValidHttpUrl function from src/utils/generic.ts. The fix in version 0.9.12 removes the vulnerable tool entirely rather than patching the URL validation logic.
View original advisory →openssl_encrypt versions prior to 1.4.9 contain a critical signature verification vulnerability in the gpg_runner.verify_detached function. The flaw arises from only checking the VALIDSIG status while ignoring REVKEYSIG, EXPKEYSIG statuses and GPG exit codes. This oversight allows attackers who possess revoked or expired signing keys to bypass signature verification entirely. Successful exploitation enables the execution of malicious plugins within the host process. The vulnerability poses a significant risk as it undermines the integrity of the plugin loading mechanism. Users are advised to upgrade to version 1.4.9 or later to remediate the issue.
View original advisory →A critical vulnerability chain has been discovered in the Avada WordPress theme, one of the most popular commercial WordPress themes. The flaw allows unauthenticated attackers to execute arbitrary PHP code on the server without any user interaction, qualifying it as a zero-click remote code execution (RCE) vulnerability. The attack requires no authentication, significantly raising its risk level and potential for mass exploitation. WordPress sites using the Avada theme are at risk of full server compromise. The vulnerability chain suggests multiple flaws are combined to achieve code execution. Site administrators using Avada are strongly urged to apply patches or mitigations immediately. The critical nature of this vulnerability, combined with the wide adoption of the Avada theme, could expose a large number of websites to attack.
View original advisory →A newly disclosed hardware attack named GPUThor targets NVIDIA GPUs by exploiting Rowhammer-style bit-flip vulnerabilities to bypass Error-Correcting Code (ECC) memory protections. The attack can enable denial-of-service (DoS) conditions as well as root-level privilege escalation on affected systems. ECC is typically considered a robust defense against Rowhammer attacks in DRAM, but GPUThor demonstrates it can be defeated in GPU memory contexts. This represents a significant threat to cloud and high-performance computing environments where NVIDIA GPUs are widely deployed. The attack highlights the growing security risks associated with GPU hardware, which is increasingly used in sensitive workloads including AI and cryptographic operations. Successful exploitation could allow an attacker to gain elevated privileges or destabilize GPU-dependent systems.
View original advisory →Medical technology company Boston Scientific has been targeted by a cyberattack that disrupted some of its IT systems and caused operational disruptions globally. The attack impacted the company's operations across multiple regions. Boston Scientific is a major medical device manufacturer, making this incident significant for the healthcare and medical technology sector. The full scope and nature of the attack, including whether ransomware was involved or data was exfiltrated, is not fully detailed in the article. The incident highlights ongoing cybersecurity threats targeting critical healthcare and medical technology infrastructure.
View original advisory →Attackers are actively targeting a chained exploit involving two Microsoft SharePoint vulnerabilities that together allow remote code execution on unpatched servers. The threat intelligence firm Defused reported the active exploitation attempts. A proof-of-concept (PoC) exploit is available, significantly lowering the barrier for attackers. The vulnerability chain enables arbitrary code execution without requiring high privileges. Organizations running unpatched SharePoint servers are at significant risk. The chained nature of the exploit makes it particularly dangerous as it bypasses individual mitigations. Microsoft has released patches and administrators are urged to apply them immediately. This follows a broader trend of SharePoint vulnerabilities being targeted by sophisticated threat actors.
View original advisory →Cybersecurity researchers have uncovered a new adversary-in-the-middle (AitM) phishing toolkit called NovaCookies, offered as a subscription-based service at $320/month. The toolkit operates as a proxy to redirect Microsoft 365 sign-in flows, capturing authenticated session cookies in the process. NovaCookies campaigns abuse legitimate Docusign notification emails to lend credibility to the phishing attempts, making detection more difficult for end users. By intercepting real authentication sessions, attackers can bypass multi-factor authentication (MFA) protections. The platform was analyzed and reported by Island, who shared findings with The Hacker News ahead of publication. This represents a growing trend of phishing-as-a-service (PhaaS) platforms targeting enterprise identity infrastructure. The abuse of genuine third-party service notifications highlights the sophistication of modern phishing campaigns.
View original advisory →