A flaw has been found in WuzhiCMS up to 4.1.0. This affects an unknown function of the file /index.php?m=member&v=Login of the component Login. This manipulation of the argument forward causes open redirect. The attack can be initiated remotely. The exploit has been published and may be used. The only sanitization is remove_xss(), an XSS keyword/entity scrubber. The vendor was contacted early about this disclosure but did not respond in any way.
CVE-2026-91203 is a symlink race condition vulnerability in the cockpit-files component of Red Hat Enterprise Linux 10. It allows a local attacker with some privileges to exploit a timing issue during privileged file operations, such as changing file ownership or permissions. By manipulating directory entries and winning the race, the attacker can redirect these operations to arbitrary files, potentially causing unauthorized changes to file ownership and permissions. This can compromise system integrity and availability by altering system or application states or rendering services unusable. Exploitation requires a privileged Cockpit user to perform file operations in directories writable by the attacker and to win the timing race. The practical impact is limited to specific operational scenarios involving writable directories and privileged Cockpit sessions.
CVE-2026-91205 is a moderate severity race condition vulnerability in the cockpit-files package used by Red Hat Enterprise Linux 9 and 10, Fedora, and Red Hat In-Vehicle OS. A local unprivileged attacker with write access to a parent directory can exploit a timing window between directory creation and ownership change to replace the directory with a symbolic link. This can redirect ownership changes to arbitrary files, potentially leading to information disclosure or unauthorized modification of sensitive files such as /etc/shadow. Exploitation requires local access, writable parent directory, user interaction, and winning the race condition. A fix is being developed for actively maintained releases, while deferred streams have mitigations recommended. Mitigations include avoiding privileged owner-assignment directory creation in writable directories by untrusted users and applying sticky bits to shared directories.
Obot versions up to 0.22.1 with authentication enabled are vulnerable to an OAuth dynamic client registration flaw that allows an attacker to steal API tokens via audience confusion. An unauthenticated attacker can register an OAuth client with an arbitrary redirect URI, and the authorization flow auto-completes without user consent. If a logged-in user visits a crafted authorization URL, the attacker receives an authorization code that can be exchanged for an access token carrying the victim's full group permissions. This token can be used against API endpoints the victim is authorized for, enabling unauthorized access until token revocation. The vulnerability is fixed in version 0.23.0 with added consent screens, token audience restrictions, and validation.
Kyoo is a self-hosted media server focused on movies, series, and anime. Prior to 5.1.0, an unauthenticated attacker could initiate the OIDC login flow with an attacker-controlled redirectUrl. The login handling in auth/oidc.go stored that URL with the opaque login state, and /auth/oidc/logged/{provider} appended the provider, token, and error values before redirecting the victim's browser without validating the destination. Because the one-use token was not bound to the browser session that initiated login, an attacker who induced a victim to complete OIDC authentication could capture the token at the attacker-controlled destination and exchange it through /auth/oidc/callback/{provider} for the victim's Kyoo session. This issue is fixed in version 5.1.0.
Vvveb is a powerful and easy to use CMS with page builder to build websites, blogs or ecommerce stores. Prior to 1.0.8.5, getThemeFolder() in admin/controller/editor/revisions.php returns the attacker-controlled theme parameter without sanitization, and backupFolder() concatenates it beneath DIR_THEMES before editor/revisions/load or editor/revisions/delete operates on a named .html file. sanitizeBackupFileName() strips traversal characters only from the separate file parameter and does not protect the theme directory component. An authenticated user with the default Editor role and editor/* permission can submit traversal sequences that redirect file_get_contents() or unlink() to a reachable backup subdirectory outside the web root. A valid admin session and CSRF token are required, the read is limited to .html files in backup directories, and deletion additionally requires filesystem write permission. This can disclose sensitive exported site content or remove backup data. This issue is fixed in version 1.0.8.5.
CloudTAK is a browser-based Common Operating Picture and situational awareness tool compatible with TAK. Prior to 13.22.1, the authenticated PUT /api/basemap endpoint passes an attacker-controlled URL through importBasemapURL() in api/routes/basemap.ts to fetch(url) without resolved-address classification or redirect revalidation. BasemapProtocol.isValidURL in api/lib/interface-basemap.ts checks only the HTTP or HTTPS scheme and is not applied on the vulnerable import path. Direct internal addresses, alternate IP encodings, and redirects to internal addresses can reach cloud metadata, loopback, private, and CGNAT HTTP services. The OptionalTileJSON response reflects fields including name, attribution, and tiles[0] to the caller, making the request forgery full-read rather than blind and enabling cloud credential theft and internal service disclosure. This issue is fixed in version 13.22.1.
mport is the MidnightBSD Package Manager. Prior to 2.7.8, privileged package fetch and cache-cleaning operations used race-prone path handling across libmport/fetch.c, libmport/clean.c, libmport/util.c, libmport/bundle_read_install_pkg.c, libmport/delete_primative.c, and libexec/mport.create/mport.create.c. A local attacker with write access to a participating package cache or staging path could race path checks and replacement operations to redirect package downloads, cleanup, or install-related side effects outside the intended cache. The affected lifecycle helper paths also used shell-form invocation, increasing command-line interpretation risk during privileged helper execution. This issue is fixed in version 2.7.8.
Use of HTTP Request With Sensitive Query String vulnerability in team-alembic AshAuthenticationPhoenix allows someone able to read access logs, proxy logs or browser history to recover a single-use sign-in token and authenticate as its owner.
After a successful password sign-in, AshAuthentication.Phoenix.Components.Password.SignInForm builds the sign_in_with_token path with the freshly issued user.__metadata__.token as a query parameter and redirects the browser to it with a GET. The token therefore travels in the request line, where web servers, reverse proxies, request telemetry and the browser's own history record it, all of which outlive the request and are ordinarily less protected than session storage. The redirect destination is restricted to a local path, so this is not an open redirect; the exposure is the retention of a live credential.
This issue affects ash_authentication_phoenix: from 1.7.0 before 2.17.4 and from 3.0.0-rc.0 before 3.0.0-rc.11; ash_authentication: from 3.10.5 before 4.15.0 and from 5.0.0-rc.0 before 5.0.0-rc.14.
Nodemailer versions from 6.9.16 up to but not including 9.1.0 contain a parsing flaw in handling RFC 5322 comments within email addresses. This flaw causes the library to incorrectly concatenate domain parts when comments are placed immediately before non-whitespace characters, leading to misinterpretation of the recipient domain. An attacker can exploit this to redirect emails to a domain they control by crafting addresses like user@good-corp.com(x)evil.com, which Nodemailer reads as good-corp.comevil.com. This vulnerability is fixed in version 9.1.0.
Craft CMS versions 4.8.0 through 4.18.5 and 5.0.0 through 5.10.12 contain a vulnerability where an authenticated non-administrator user can execute arbitrary operating system commands. This occurs due to improper binding of HMAC signatures to their purpose, allowing an attacker to reuse a signed cookie in a redirect parameter that is rendered as an unsandboxed Twig template. The issue requires password authentication without 2FA and availability of PHP system(). The vulnerability is fixed in versions 4.18.6 and 5.10.13.
RMCP is an official Rust SDK for the Model Context Protocol. Prior to 2.1.0, the rmcp crate's StreamableHttpClientTransport in crates/rmcp/src/transport/common/reqwest/streamable_http_client.rs builds its default_http_client with reqwest's automatic redirect policy and applies caller-supplied values from StreamableHttpClientTransportConfig.custom_headers without marking them as sensitive. When a malicious or compromised MCP endpoint returns a cross-origin 307 or 308 redirect, reqwest follows the redirect and apply_custom_headers causes custom API keys or authentication tokens to be replayed to the new origin, where an attacker can capture and reuse them. The separate auth_header path is not affected because it uses the standard Authorization header, which reqwest strips on cross-origin redirects. This issue is fixed in version 2.1.0.
BragJack is a security research discovery demonstrating how a single browser extension can hijack internal browser agents across five Chromium-based browsers: Chrome, Comet, Edge, Opera Neon, and Claude in Chrome. The attack leverages the declarativeNetRequest (DNR) API to weaken security headers and redirect JavaScript resources, enabling code execution within privileged browser contexts. This allows access to sensitive capabilities such as local file access, browsing history, camera and microphone, and screenshots without user interaction. The research resulted in two CVEs and $20,000 in bounty rewards from major vendors. The vulnerabilities primarily exploit weaknesses in how these browsers handle network requests and security policies within embedded web views. No patch or remediation status is provided in the source data.
Red Hat has issued a security update for the Red Hat build of Keycloak 26.4.16 images used in OpenShift Container Platform. The update addresses multiple vulnerabilities including privilege escalation via impersonation roles, information leakage through SAML Redirect DEFLATE helpers, account lockout caused by username collisions, incomplete fixes for filesystem path probing and policy enforcer bypass, TLS hostname verification bypass, and an unauthenticated denial-of-service via unbounded locale caching. These issues affect the Red Hat build of Keycloak versions from 26.4.0 up to and including 26.4.16. The update is important to maintain the security and integrity of authentication services in OpenShift environments.
A flaw was found in the SAML Redirect Binding implementation of Keycloak, an open-source identity and access management solution. The issue occurs because the custom DEFLATE compression and decompression helpers fail to release native zlib memory after use. An unauthenticated attacker can exploit this by sending repeated malformed SAML requests, leading to native memory exhaustion and a denial of service.
CVE-2026-92216 is an open redirect vulnerability in the a2ui-project's a2ui component, specifically affecting the openUrl function in the Binder component. This vulnerability exists in versions 0.10.0 through 0.10.7 and allows remote attackers to manipulate URLs to redirect users to unintended destinations. The issue was reported to the project early, but no response or fix has been issued yet. The vulnerability has a medium severity rating with a CVSS score of 5.3.