Overview of incoming advisories.
2018 results found
FastChat, an open-source LLM serving framework, contains a critical authentication bypass vulnerability in its /register_worker endpoint. Unauthenticated attackers can exploit this flaw to register arbitrary worker addresses without any authentication. The vulnerability enables server-side request forgery (SSRF) by allowing attackers to register malicious workers under legitimate victim model names. This facilitates interception of sensitive user data including prompts, images, and model responses. Attackers can also leverage the vulnerability to probe and scan internal network ports across the worker mesh infrastructure. The issue is present in at least version v0.2.36 of FastChat as referenced in the controller.py source file. The vulnerability poses significant risks to AI/ML deployments using FastChat for model serving, potentially exposing confidential user interactions and internal network topology.
View original advisory →A stack-based buffer overflow vulnerability has been identified in FreeIPMI versions prior to 1.6.19. The flaw exists in the ipmi-oem component, specifically within the _output_dell_system_info_cmc_info function located in ipmi-oem/ipmi-oem-dell.c. The vulnerability is triggered via the cmc-info subcommand of the dell get-system-info operation. Stack-based buffer overflows can potentially allow attackers to execute arbitrary code or cause a denial of service. The vulnerability has been assigned CVE-2026-85507 and is tracked by NVD. A patched version, FreeIPMI 1.6.19, has been released and is available via the GNU FTP server. The issue was disclosed on the oss-security mailing list in August 2026. Users are advised to upgrade to FreeIPMI 1.6.19 or later to mitigate the risk.
View original advisory →The ACPT (Premium) plugin for WordPress contains a critical privilege escalation vulnerability affecting all versions up to and including 2.0.66. The flaw resides in the submit() function, which lacks proper authorization checks, allowing unauthenticated attackers to manipulate the target user ID in form submissions. By exploiting this, an attacker can overwrite any WordPress user's email address and password, including administrator accounts, effectively taking full control. Exploitation is possible when a public ACPT user form that permits anonymous submissions is present on the site. This vulnerability enables complete account takeover without any prior authentication. The issue has been documented by both NVD and Wordfence threat intelligence. Site administrators are urged to update beyond version 2.0.66 immediately to mitigate risk.
View original advisory →A SQL injection vulnerability has been identified in the light0011 CMS project at specific commits. The flaw resides in the ChapterModel::searchChapter function within App/Home/Controller/ChapterController.class.php, where the 'content' argument is not properly sanitized by the Query Builder component. Remote attackers can exploit this vulnerability without authentication. A public exploit has been released, increasing the risk of active exploitation. The product does not follow versioning, making it impossible to identify affected or unaffected releases. The project maintainer was notified via a GitHub issue but has not responded. This vulnerability poses a significant risk to any deployments of this CMS.
View original advisory →A SQL injection vulnerability has been identified in the light0011 CMS at specific commit hashes (c774dce31c6df0055568a8d5c53d964d99be199d and f72cf46f601efb2a0618c3814cc2f61380b38930). The flaw resides in the ChapterModel::searchChapter function within App/Home/Controller/ChapterController.class.php, specifically in the Query Builder component. Attackers can manipulate the 'content' argument to perform SQL injection attacks remotely. A public exploit has already been released, increasing the risk of active exploitation. The product does not use versioning, making it impossible to identify specific affected or unaffected releases. The project maintainer was notified via a GitHub issue report but has not yet responded. This vulnerability poses a significant risk to any deployment of this CMS.
View original advisory →The Divi Ajax Filter plugin for WordPress contains a Local File Inclusion (LFI) vulnerability affecting all versions up to and including 5.1.2. The flaw exists in the 'custom_loop_template' parameter and allows unauthenticated attackers to include and execute arbitrary PHP files on the server. Successful exploitation can lead to access control bypass, sensitive data exposure, or full remote code execution if PHP files can be uploaded. The vulnerability is conditionally exploitable only when the 'loop_templates' parameter is set to 'custom-template'. No authentication is required, making this a significant risk for any WordPress site running the affected plugin versions. A patch or updated version is expected as referenced in the Divi Engine changelog. The issue has been documented by both NVD and Wordfence threat intelligence.
View original advisory →FreeIPMI versions before 1.6.19 contain a stack-based buffer overflow vulnerability in the function _ipmi_sel_oem_fujitsu_get_sel_entry_long_text located in libfreeipmi/sel/ipmi-sel-string-fujitsu-irmc-common.c. The vulnerability is triggered via malformed Fujitsu SEL long-text responses, which can cause memory corruption on the stack. This could potentially allow an attacker to execute arbitrary code or crash the affected application. The issue specifically affects systems using Fujitsu iRMC OEM SEL entries. A fix has been released in FreeIPMI version 1.6.19. Users are advised to update to the patched version immediately. The vulnerability was disclosed via the oss-security mailing list and the fix is available via the official GNU FTP server.
View original advisory →Aim version 3.29.1's remote tracking server contains a critical authentication bypass vulnerability. The server fails to authenticate incoming requests and dispatches arbitrary methods via Python's getattr without any allowlist validation. Unauthenticated attackers can exploit this flaw to register clients, instantiate Repo resources, and invoke arbitrary methods on the server. This allows attackers to read experiment data or delete runs without any credentials. The vulnerability resides in the transport layer components, specifically server.py and tracking.py. The issue has been documented in the Aim GitHub repository and reported via VulnCheck advisories. This represents a significant risk for any organization exposing the Aim remote tracking server to untrusted networks.
View original advisory →Blinko version 1.8.7 contains an Insecure Direct Object Reference (IDOR) vulnerability affecting multiple tRPC procedures including message.list, message.update, message.delete, message.clearAfter, and conversation.clearMessages. While these endpoints require authentication, they fail to verify that the requested resource belongs to the authenticated user, instead trusting caller-supplied conversation or message IDs directly. An attacker with a valid account can exploit this by enumerating sequential integer IDs to access, modify, or delete other users' AI chat histories and conversations. The vulnerability exposes sensitive user data and allows unauthorized data manipulation across user boundaries. The issue was identified in server/routerTrpc/message.ts and server/routerTrpc/conversation.ts source files. A fix appears to have been introduced in version 1.8.8 based on the patched file references in GitHub.
View original advisory →A SQL injection vulnerability has been identified in code-projects Doctor Appointment System version 1.0. The flaw exists in the /contactus.php file, where manipulation of the 'firstname' argument allows SQL injection attacks. The vulnerability can be exploited remotely without requiring local access. A public exploit has already been published and is available for use by threat actors. This makes the vulnerability particularly dangerous as it lowers the bar for exploitation. The affected product is a web-based doctor appointment scheduling system. Organizations using this software should apply patches or mitigations immediately given the public availability of the exploit.
View original advisory →Plandex version 2.2.1 contains a path traversal vulnerability in the ApplyFiles function that allows attackers to write files outside the intended project directory. The flaw can be exploited by influencing AI model output through poisoned repository files or attacker-controlled context inputs. Successful exploitation enables writing to arbitrary locations such as shell rc files or cron job configurations. This can lead to arbitrary code execution on the affected system. The vulnerability is tracked as CVE-2026-85690 and has been documented by VulnCheck. The affected code resides in app/cli/lib/apply.go within the Plandex CLI. The issue has been reported via GitHub and is publicly disclosed. Users of Plandex 2.2.1 are advised to review available patches or mitigations immediately.
View original advisory →CVE-2026-85608 affects Douyin_TikTok_Download_API through version 4.1.2, exposing a server-side request forgery (SSRF) vulnerability in two endpoints: /api/download and /api/hybrid/video_data. Unauthenticated attackers can exploit this flaw by supplying an arbitrary URL via the 'url' query parameter, causing the server to fetch the specified resource on their behalf. This allows attackers to reach internal services, including cloud metadata endpoints such as AWS IMDSv1, which can expose sensitive credentials and configuration data. Sensitive information can be leaked through error messages returned in the response body. No authentication is required to exploit this vulnerability, significantly increasing its risk surface. The affected code is identified in the crawlers/douyin/web/utils.py file in the repository. The vulnerability is documented in the project's GitHub issue tracker and in a VulnCheck advisory. Organizations using this API in cloud-hosted or internally networked environments are particularly at risk.
View original advisory →SmartIT Desktop Manager, developed by Lightstar, contains a Use of Hard-coded Credentials vulnerability identified as CVE-2026-85148. The vulnerability allows unauthenticated remote attackers to exploit a fixed, embedded password to gain unauthorized remote access to user hosts. No authentication is required to leverage this flaw, making it particularly dangerous for affected deployments. The issue is classified as high severity given the ease of exploitation and the potential for full remote access. The vulnerability has been reported via Taiwan CERT (TWCERT) and published on the NVD. Organizations using SmartIT Desktop Manager by Lightstar should apply patches or mitigations immediately. Hard-coded credentials are a well-known and critical class of vulnerability that bypasses standard access controls entirely.
View original advisory →A vulnerability has been identified in jofpin trape 2.0, an open-source tool used for tracking and phishing. The flaw exists in the file core/user.py, where manipulation of the vId/id arguments leads to an authorization bypass. Remote exploitation is possible without authentication, making this a significant security risk. A public exploit has already been released, increasing the likelihood of active exploitation. The project maintainer was notified via a GitHub issue but has not responded or issued a patch. The vulnerability is classified as an authorization bypass, allowing attackers to potentially access restricted resources or functionality. Given the public availability of the exploit and the lack of vendor response, users of trape 2.0 are at elevated risk.
View original advisory →A SQL injection vulnerability has been identified in code-projects Hospital Information System version 1.0. The vulnerability exists in the viewReq function within the viewReq.php file, where manipulation of the 'ID' argument allows SQL injection attacks. The flaw can be exploited remotely without requiring physical access to the target system. A public exploit is already available, increasing the risk of active exploitation. The vulnerability affects the integrity and confidentiality of the underlying database. Given the healthcare context, sensitive patient data may be at risk. The CVE identifier assigned is CVE-2026-85398, and it has been documented across multiple security databases including NVD and VulDB.
View original advisory →A SQL injection vulnerability has been identified in code-projects Vehicle Management System 1.0. The vulnerability resides in an unknown function within the file /busprofile.php, where manipulation of the 'busid' argument allows for SQL injection attacks. The vulnerability can be exploited remotely without requiring physical access to the system. A public exploit is already available, increasing the risk of active exploitation. The affected product is a web-based vehicle management application. The vulnerability has been assigned CVE-2026-85516 and is tracked in VulDB as entry 398680. No patch or workaround details are mentioned in the article. Organizations using this software should take immediate action to mitigate exposure.
View original advisory →CVE-2026-9317 is a critical missing authentication vulnerability in Nango versions before 0.71.6, affecting the runner tRPC server. Unauthenticated attackers with network access to the runner port can invoke the exposed 'start' procedure without credentials. The vulnerability stems from an unenforced RUNNER_SECRET_KEY environment variable, allowing attackers to bypass authentication entirely. Successful exploitation enables remote code execution of arbitrary JavaScript within the runner process. The fix was introduced in Nango v0.71.6 via a commit to the NangoHQ GitHub repository. Organizations running Nango in environments where the runner port is network-accessible are at significant risk. Users should upgrade to v0.71.6 or later immediately to remediate this vulnerability.
View original advisory →Postgres MCP Pro version 0.3.0 contains a security vulnerability that allows attackers to bypass restricted-mode protections. The flaw exists because function-name validation is not applied to RangeFunction nodes in SQL FROM clauses. Attackers can exploit this by using FROM-clause syntax to invoke sensitive functions such as pg_read_file, enabling arbitrary file reads on the server. This bypass undermines the intended security boundaries of restricted mode. The vulnerability is tracked as CVE-2026-85620 and was identified in the safe_sql.py component of the project. It poses a significant risk to environments relying on Postgres MCP Pro's restricted mode for access control. A fix would require extending function-name validation to cover RangeFunction nodes in FROM clauses. The issue has been reported via GitHub and documented by VulnCheck.
View original advisory →A SQL injection vulnerability has been identified in code-projects Doctor Appointment System version 1.0. The flaw exists in the file /patient/booking.php, where manipulation of the doc_id argument allows an attacker to perform SQL injection attacks. The vulnerability can be exploited remotely without requiring physical access to the target system. A public exploit is already available, increasing the risk of active exploitation. The affected product is a web-based doctor appointment management system. The vulnerability has been assigned CVE-2026-85402 and is tracked in multiple vulnerability databases including NVD and VulDB. Organizations using this software should apply patches or mitigations immediately given the public availability of the exploit.
View original advisory →Acunetix version 25.11.251107123 for Windows contains a local privilege escalation vulnerability in its Web Vulnerability Scanning Engine (wvsc.exe). The vulnerability stems from a missing hardcoded directory path for OpenSSL-related files. A low-privileged local attacker can exploit this by creating the missing directory and placing a malicious file at the expected path. The SYSTEM-level wvsc.exe process will then load and execute the malicious file, resulting in full privilege escalation to SYSTEM. This represents a significant security risk as it allows attackers to gain the highest level of Windows system privileges. A proof-of-concept has been published and the vulnerability has been disclosed via full disclosure mailing lists.
View original advisory →