Parse Server versions 2.5.0 before 8.6.121 and 9.0.0 before 9.11.1-alpha.1 contain a vulnerability where a failure to serialize an object after a beforeSave trigger can cause the server process to terminate. This occurs when a remote attacker updates an object with a dot notation key below a non-object field, such as a timestamp, leading to an uncaught exception and server crash.
Parse Server versions before 8.6.122 and 9.0.0 before 9.11.1-alpha.2 contain an authorization flaw that allows clients with write permissions to modify certain fields using dot notation keys. This bypasses beforeSave trigger protections on fields such as Object, Array, GeoPoint, Polygon, Bytes (on MongoDB), and ACL during create and update operations, potentially circumventing Cloud Code authorization checks.
SiYuan versions 3.8.0 through 3.8.9 contain an information disclosure vulnerability in the LAN sync feature. An unauthenticated attacker on the local network can impersonate a peer and cause the client to send sensitive device metadata, including sync scope ID, device name, OS, app version, and TLS client certificate, before verifying the server's authenticity. This vulnerability is fixed in version 3.8.10.
SiYuan versions 3.8.0 through 3.8.9 contain an incorrect authorization vulnerability in the serveAssets route. This flaw allows unauthenticated attackers to read password-protected document assets by exploiting name collisions. Specifically, if a visible document references an asset with the same relative name, an attacker can retrieve the protected file's contents via a GET request on public-publish instances where Basic Auth is disabled. The issue is fixed in version 3.8.10.
SiYuan versions 3.6.0 before 3.6.2 and 3.7.3 before 3.8.10 have a missing authorization vulnerability in the /api/block/getDOMText endpoint. This flaw allows anonymous users to retrieve cached OCR text of unpublished, non-encrypted assets when publish authentication is disabled, due to lack of access checks on referenced image paths.
SiYuan versions 3.8.0 through 3.8.9 contain an incorrect authorization vulnerability that allows anonymous attackers to access confidential indexed asset text. Attackers can exploit this by requesting specific API endpoints on public host documents to retrieve sensitive file contents. The issue is fixed in version 3.8.10.
SiYuan versions 3.6.0 up to but not including 3.8.10 contain an information disclosure vulnerability. Authenticated users with Reader privileges can exploit block-reference footnote resolution in Markdown export to access protected documents. This includes documents that are hidden, disabled, or password-protected, allowing disclosure of their body text without needing the publish password. The issue is fixed in version 3.8.10.
SiYuan note-taking software versions prior to 3.8.10 contain an authentication bypass vulnerability. This flaw allows unauthenticated attackers to gain Administrator privileges by exploiting the handling of the X-Forwarded-For header when requests are proxied through a trusted loopback reverse proxy. By sending requests with the Host header set to localhost and X-Forwarded-For set to 127.0.0.1, attackers can access private note history via the /history/*path endpoint if no lock-screen password is configured. The issue is fixed in version 3.8.10.
A stored cross-site scripting (XSS) vulnerability exists in the Grav Form plugin versions from 1.1.0 up to but not including 9.1.40. The vulnerability arises because the Form::uploadFiles() function only sanitizes files with a last extension of .svg, allowing files with extensions like .svg.foo to bypass sanitization. This can lead to script execution in an administrator's session, potentially enabling an attacker to create a super-admin account on Grav 2.x with Admin2.
Grav API plugin versions from 1.0.16 up to but not including 1.1.11 contain an authorization bypass vulnerability. This flaw allows scoped API key holders to override page permission rules because the system checks the owning account's authority instead of the API key's scopes. Attackers with keys minted on super-admin accounts can perform unauthorized edits, deletions, or reads on pages that normally deny those actions. Similarly, media-scoped keys can modify media on pages that are intended to be read-only. The vulnerability has a medium severity score of 5.3 and was fixed in version 1.1.11.
Grav API plugin versions from 1.0.0 up to but not including 1.1.4 contain a remote code execution vulnerability. This flaw allows users with media and page write permissions to upload files into a theme's Twig templates folder via the blueprint-upload endpoint. These uploaded files can then be included and executed within sandboxed page Twig templates, resulting in command execution as the web-server user. The vulnerability is fixed in version 1.1.4.
Admidio versions from 5.0.0 up to but not including 5.0.18 contain a path traversal vulnerability in the DocumentsService::renameFile() function. Authenticated users with folder upload permission can exploit this flaw by supplying crafted adm_new_name values containing '../' sequences, allowing them to move files outside the intended adm_my_files directory. This enables writing files and creating directories in arbitrary writable locations, including the web root. The vulnerability is rated high severity with a CVSS 4.0 score of 7.1. A fix is available in version 5.0.18.
Admidio versions from 5.0.17 up to but not including 5.0.18 contain a stored cross-site scripting (XSS) vulnerability. Authenticated users can inject malicious JavaScript into the STREET profile field, which is not properly neutralized during web page generation. This injected script executes when role managers hover over the address icon, exploiting an unencoded title attribute in members_assignment_data.php. The vulnerability is fixed in version 5.0.18.
Admidio versions from 5.0.17 up to but not including 5.0.18 have a missing authorization vulnerability in the announcements module. Authenticated regular users can exploit the CKEditor upload handler to upload images without proper authorization. This allows attackers to store images under adm_my_files/announcements/images and retrieve them via show_image.php, effectively abusing the application as an image hosting service.
Admidio versions prior to 5.0.18 have a session management vulnerability where sessions are not properly invalidated after a password reset. This allows attackers who already possess a victim's session or remember-me cookie to maintain access to the account even after the victim changes their password.
Admidio versions 5.0.0 up to but not including 5.0.18 contain a path traversal vulnerability in the Documents & Files renameFile() action. Authenticated users with folder upload rights can exploit this flaw by using ../ sequences in the adm_new_name parameter to move files outside the intended directory. This allows attackers to relocate malicious SVG files containing scripts to the web root, enabling stored cross-site scripting attacks against administrators that can lead to account takeover.
Admidio versions from 5.0.16 up to but not including 5.0.18 contain a stored cross-site scripting (XSS) vulnerability. This flaw allows users with folder upload permissions to upload SVG files containing malicious JavaScript. The vulnerable function serves these SVG files inline without appropriate security headers such as nosniff or Content Security Policy (CSP), enabling script execution in the context of the Admidio origin for all viewers, including unauthenticated visitors. The issue is fixed in version 5.0.18.
Admidio versions 5.0.0 through 5.0.17 contain an improper authorization vulnerability in the profile management module. Authenticated Group Leaders can bypass field-level profile restrictions to read and modify restricted profile fields by exploiting specific request modes. This issue is fixed in version 5.0.18.
Admidio versions 5.0.17 up to but not including 5.0.18 contain an authorization bypass vulnerability in the Lists module. Authenticated users with access to the Lists module can use a list UUID to access list configurations belonging to other users within the same organization. This allows reading of list names, selected profile fields, sort orders, and filter conditions without proper ownership or privacy checks.
Admidio versions 5.0.0 through 5.0.17 contain a stored cross-site scripting (XSS) vulnerability. This flaw allows role administrators to inject malicious JavaScript code into role names, which are rendered without proper encoding in a profile page attribute. When a full administrator views the affected profile page, the injected script can execute, enabling theft of the CSRF token and unauthorized assignment of the Administrator role. The vulnerability is fixed in version 5.0.18.