Mave is a Tel Aviv-based startup founded by former Unit 8200 officers that uses AI agents to automate and enhance security operations by investigating alerts across existing security stacks. The company aims to address the challenge of overwhelming data and slow response times in enterprise security teams by leveraging AI to keep pace with AI-assisted attackers. Mave's solution integrates with existing security tools to provide near real-time context-based reasoning, reducing alert fatigue and improving operational efficiency. The startup emphasizes controlled AI autonomy with approval gates and audit trails to mitigate risks associated with automated actions. Currently, Mave does not offer full automation but focuses on augmenting human decision-making in security operations. The company recently raised $34 million to accelerate AI research and expand its product offerings and global presence.
CVE-2026-116050 is a SQL injection vulnerability in the krupalshah Employee Recruitment System affecting the /source/registerform3.php file's pemail parameter. The vulnerability allows remote attackers to manipulate SQL queries. The product uses continuous delivery with rolling releases, so no specific affected or fixed versions are available. The vendor did not respond to the disclosure and the affected product is no longer supported. The vulnerability has a medium severity score of 6.9 and public exploit details have been disclosed.
CVE-2026-116049 is a SQL injection vulnerability in the krupalshah EmployeeRecruitmentSystem affecting the mysql_query function in /source/registerform2.php. The vulnerability allows remote attackers to manipulate several input parameters, potentially leading to SQL injection. The product uses rolling releases, so specific affected or fixed versions are not available. The vendor did not respond to disclosure and no patch or remediation is currently known. This vulnerability affects unsupported versions of the product.
A malicious file identified by the SHA-256 hash a151d3f4f2422531f30a843ffb35479596c86722bb103bdf8591105687f9b125 was observed by OffSeq Mirage honeypot sensors on 2026-10-25. The file is associated with attacker activity involving the MITRE ATT&CK technique T1105 (Ingress Tool Transfer). This hash fingerprints a specific malicious payload that could be a dropper, web shell, miner, or post-exploitation tool. The activity was seen originating from attacker infrastructure in KP. No known exploits in the wild or specific affected software versions are reported.
CVE-2026-116048 is a SQL injection vulnerability in the krupalshah EmployeeRecruitmentSystem affecting the mysql_query function in /source/registerform1.php. The vulnerability allows remote attackers to manipulate multiple input parameters leading to SQL injection. The product uses a rolling release model, so no specific affected or fixed versions are available. The vendor did not respond to disclosure and the affected product is no longer supported. The vulnerability has a CVSS 4.0 score of 6.9 (medium severity).
CVE-2026-116047 is a cross-site scripting (XSS) vulnerability in Privoce Vocechat version 0.9.121. The flaw exists in the processing of the 'welcome' query parameter within the widget.html component. This vulnerability can be exploited remotely and requires user interaction. The vendor has not responded to the disclosure, and no patch or remediation information is currently available.
A security researcher discovered a method to bypass the AWS MCP Python sandbox restrictions in the aws___run_script tool, which executes user-supplied Python code in a restricted AWS Lambda environment. The sandbox uses AST validation and runtime proxy objects to block dangerous operations, including attribute access and imports. However, the researcher found that Python's structural pattern matching (match/case) can implicitly read attributes listed in __match_args__, bypassing the sandbox's attribute access restrictions. By assigning __match_args__ to sensitive dunder attributes and constructing attribute names at runtime, the researcher achieved sandbox escape and system command execution within the Lambda environment.
CVE-2026-116046 is a path traversal vulnerability in the eventcatalog sdk affecting versions up to 2.29.0. The flaw exists in the Custom Documentation Helpers component, specifically in the functions getCustomDoc, writeCustomDoc, and rmCustomDoc. It allows manipulation of the options.path argument to traverse directories on the local host. The issue is resolved by upgrading to version 2.30.0.
CVE-2026-116045 is a path traversal vulnerability in 1999AZZAR designer-mcp up to version 2.1.0. It affects the get_reference function in the MCP Interface component, allowing remote attackers to manipulate the argument Name to perform path traversal. The vulnerability has a medium severity with a CVSS score of 5.3. A patch identified by commit c8c2951dcbed9b1710a745993f028317f10a3f10 is available and applying it is advised.
The npm package '@eforms/forms-api' version 2.0.0-SNAPSHOT has been identified as malicious. The package is flagged because it communicates with a domain linked to malicious activity. There is no CVSS score available, and no known exploits in the wild have been reported. No patch or remediation information is provided.
The Duplicator WordPress plugin before 5.0.0 does not restrict its one-click upgrade routine to users allowed to install Duplicator WordPress plugin before 5.0.0, and additionally exposes the routine without authentication, allowing an administrator of a multisite child site who is not a network administrator to install and run arbitrary PHP in the network-wide shared Duplicator WordPress plugin before 5.0.0 directory.
The Datawiza Proxy Auth Plugin for WordPress before version 1.1.1 has an improper privilege management vulnerability. It fails to verify that user-identity and role HTTP headers originate from a trusted reverse proxy. This flaw allows unauthenticated attackers to create and authenticate as WordPress administrators.
The Qi Addons For Elementor WordPress plugin before version 1.11.2 contains a vulnerability where it does not properly sanitize a widget's heading text after decoding HTML entities. This flaw allows users with Contributor role or higher to embed live HTML forms, including password fields that submit data to external hosts, into posts where WordPress normally strips such content.
The InstaWP Connect WordPress plugin before 0.1.4.2 does not perform a capability check on its InstaWP Connect WordPress plugin before 0.1.4.2-settings update action, verifying only a nonce, allowing an authenticated user holding a low-privileged role that an administrator has granted access to the InstaWP Connect WordPress plugin before 0.1.4.2's sync feature to overwrite the InstaWP Connect WordPress plugin before 0.1.4.2's settings, including the setting controlling which roles may access that feature. The role grant is not enabled by default.
The WPVR WordPress plugin versions 8.3.0 up to but not including 9.0.3 have an authorization bypass vulnerability. This flaw allows users with contributor-level access or higher to update virtual tours created by other users without proper ownership checks. As a result, unauthorized modification, reconfiguration, and publishing of other users' virtual tours is possible.
The WPVR WordPress plugin versions 9.0.0 through 9.0.2 contain an authorization bypass vulnerability. This flaw allows users with contributor-level access or higher to read and modify virtual tours created by other users due to missing object-level authorization checks. The issue is fixed starting from version 9.0.3.
A missing authorization vulnerability in the WPVR WordPress plugin before version 9.1.3 allows users with the Contributor role and above to create and publish tour content via the setup wizard. This flaw permits these users to set arbitrary post metadata that they would not normally be authorized to create or publish.
The GiveWP WordPress plugin before 4.19.0.1 does not perform a capability check when handling donor address changes, allowing users who have been granted donor-report (view) access but not donor-edit access to add, modify, or delete the stored address of any donor on the site.
GiveWP WordPress plugin versions before 4.19.0.1 contain an authorization flaw that allows users with a custom fundraising role to edit, publish, and read pages owned by other users, including administrators. This occurs because the plugin does not correctly determine which page a page-capability request applies to.
The GiveWP WordPress plugin before 4.19.0.1 does not properly sanitise and escape one of its donation form template settings before outputting it inside an HTML attribute, allowing users with a custom form-editing GiveWP WordPress plugin before 4.19.0.1 role to inject arbitrary web scripts that execute in the browser of any visitor or administrator who opens the affected donation form.