When authentication caching is enabled on DefaultLdapRealm, after a user logs in successfully once, subsequent logins for that user succeed with any password. A cached authentication skips the LDAP bind, and the AllowAllCredentialsMatcher installed by default in DefaultLdapRealm accepts any credentials, allowing an attacker who knows a valid username to authenticate as that user. This issue affects all Apache Shiro versions through 3.0.0, only when using DefaultLdapRealm (or JndiLdapRealm) with authenticationCachingEnabled set to true, which is not the default.

Mitigation: Upgrade to version 3.0.1 or later, which fixes the issue by no longer installing AllowAllCredentialsMatcher by default. Alternatively, disable authentication caching on the LDAP realm (this is the default), or configure a credentials matcher, such as SimpleCredentialsMatcher, that verifies credentials against the cached authentication info.

Dokploy before 0.31.1 contains an OS command injection vulnerability in getComposeContainerCommand() that allows authenticated organization members to execute shell commands via the serviceName parameter. Attackers can store serviceName values with shell metacharacters through backup.create or backup.update tRPC procedures, executing commands as root when scheduled backups run and taking over the host via the mounted Docker socket.

C-Kermit 6.0.192 before 11.1.515 contains a buffer overflow vulnerability in tnsnenv() in ckctel.c that allows malicious Telnet servers to corrupt client memory via crafted NEW-ENVIRON SEND subnegotiation. Malicious or machine-in-the-middle servers can send oversized variable names before authentication, overflowing a 16-byte stack buffer and a miscomputed heap buffer to corrupt stack and heap memory.

C-Kermit 6.0.192 before 11.1.515 contains a buffer overflow vulnerability in tnsnenv() in ckctel.c that allows malicious Telnet servers to corrupt client memory via crafted NEW-ENVIRON SEND subnegotiation. Malicious or machine-in-the-middle servers can send oversized variable names before authentication, overflowing a 16-byte stack buffer and a miscomputed heap buffer to corrupt stack and heap memory.

C-Kermit 6.0.192 before 11.1.515 contains a buffer overflow vulnerability in tnsnenv() in ckctel.c that allows malicious Telnet servers to corrupt client memory via crafted NEW-ENVIRON SEND subnegotiation. Malicious or machine-in-the-middle servers can send oversized variable names before authentication, overflowing a 16-byte stack buffer and a miscomputed heap buffer to corrupt stack and heap memory.

C-Kermit 6.0.192 before 11.1.515 contains a buffer overflow vulnerability in tnsnenv() in ckctel.c that allows malicious Telnet servers to corrupt client memory via crafted NEW-ENVIRON SEND subnegotiation. Malicious or machine-in-the-middle servers can send oversized variable names before authentication, overflowing a 16-byte stack buffer and a miscomputed heap buffer to corrupt stack and heap memory.

C-Kermit 6.0.192 before 11.1.515 contains a buffer overflow vulnerability in tnsnenv() in ckctel.c that allows malicious Telnet servers to corrupt client memory via crafted NEW-ENVIRON SEND subnegotiation. Malicious or machine-in-the-middle servers can send oversized variable names before authentication, overflowing a 16-byte stack buffer and a miscomputed heap buffer to corrupt stack and heap memory.

C-Kermit 6.0.192 before 11.1.515 contains a buffer overflow vulnerability in tnsnenv() in ckctel.c that allows malicious Telnet servers to corrupt client memory via crafted NEW-ENVIRON SEND subnegotiation. Malicious or machine-in-the-middle servers can send oversized variable names before authentication, overflowing a 16-byte stack buffer and a miscomputed heap buffer to corrupt stack and heap memory.

Tactical RMM 0.17.0 through 1.6.1 contains an arbitrary file read in the /reporting/templates/preview/ endpoint because WeasyPrint resolves file:// links in user-supplied templatemd and embeds the files in the returned PDF. Users with report management permission can read the Django SECRETKEY, reused as the NATS password, and publish runscript commands to every managed agent.

listmonk 6.0.0 before 7.0.0-beta.2 contains an authentication bypass vulnerability in doResetPassword that skips TOTP verification after a password reset. Attackers who can complete a reset, such as by reading the victim's reset email, obtain fully authenticated sessions including Super Admin without the TOTP code.

sysPass 3.0.1 through 3.2.11 contains an authentication bypass vulnerability that allows unauthenticated attackers to reinstall an already-installed instance by POSTing to the install/install handler, which never checks isInstalled(). Attackers can overwrite config.xml, point the application at an attacker-controlled database, create an administrator account, and plant serialized PublicLink gadget chains triggered via account/viewLink to execute commands.

OpenMage Magento LTS before 20.18.2 and 21.0.0-beta1 before 21.0.0-beta3 contains a cross-site request forgery vulnerability in the admin and storefront OAuth confirm and confirmSimple actions, which accept GET requests without form-key validation. Attackers holding OAuth consumer credentials can lure a logged-in administrator or customer to a crafted confirm URL that authorizes the attacker's request token, then exchange it at /oauth/token for a persistent REST API access token bound to the victim.

OpenMage Magento LTS before 20.18.2 and 21.0.0-beta1 before 21.0.0-beta3 contains a cross-site request forgery vulnerability in the admin and storefront OAuth confirm and confirmSimple actions, which accept GET requests without form-key validation. Attackers holding OAuth consumer credentials can lure a logged-in administrator or customer to a crafted confirm URL that authorizes the attacker's request token, then exchange it at /oauth/token for a persistent REST API access token bound to the victim.

OpenMage magento-lts 20.17.0 before 20.18.2 and 21.0.0-beta1 before 21.0.0-beta3 contains an unrestricted file upload vulnerability because the general/file/protectedextensions blocklist omits phar, php8, and phps. Unauthenticated attackers can upload a .phar file through a product File custom option to a predictable media/customoptions path and request it directly to execute code on servers that map .phar to PHP and serve that directory.

OpenMage magento-lts before 20.18.2 and 21.0.0-beta pre-releases before 21.0.0-beta3 generate customer email-confirmation keys with md5(uniqid()) and customer and administrator password-reset tokens with the non-cryptographic uniqHash(). Unauthenticated attackers can brute-force the roughly 20-bit confirmation key via the unrate-limited confirmAction endpoint to log in as pending customers, or predict administrator reset tokens after recovering generator state.

OpenMage magento-lts 20.17.0 before 20.18.2 and 21.0.0-beta1 before 21.0.0-beta3 contains an unrestricted file upload vulnerability because the general/file/protectedextensions blocklist omits phar, php8, and phps. Unauthenticated attackers can upload a .phar file through a product File custom option to a predictable media/customoptions path and request it directly to execute code on servers that map .phar to PHP and serve that directory.

FrankenPHP standalone static binaries 1.8.0 before 1.13.2 bundle the PHP brotli extension statically linked with Brotli 1.2.0, whose decoder contains an out-of-bounds write (CVE-2026-111497). Unauthenticated remote attackers can send crafted data to applications that pass it to brotliuncompressadd() to crash the entire FrankenPHP process, and possibly cause memory disclosure or code execution.

FrankenPHP standalone static binaries 1.8.0 before 1.13.2 bundle the PHP brotli extension statically linked with Brotli 1.2.0, whose decoder contains an out-of-bounds write (CVE-2026-111497). Unauthenticated remote attackers can send crafted data to applications that pass it to brotliuncompressadd() to crash the entire FrankenPHP process, and possibly cause memory disclosure or code execution.

FrankenPHP before 1.13.0 contains an authentication bypass vulnerability that allows unauthenticated attackers to spoof trusted headers by sending dot-form header names that PHP rewrites to underscores. Attackers can send headers like X.Authenticated.User or X.Forwarded.For that map to $SERVER keys, spoofing user identity, client IP, or scheme.

FrankenPHP before 1.13.0 contains an authentication bypass vulnerability that allows unauthenticated attackers to spoof trusted headers by sending dot-form header names that PHP rewrites to underscores. Attackers can send headers like X.Authenticated.User or X.Forwarded.For that map to $SERVER keys, spoofing user identity, client IP, or scheme.

Zipline through 4.4.0 contains an authentication bypass vulnerability in the WebAuthn login endpoint that creates sessions for a supplied registered credential ID without verifying the assertion. Attackers who obtain a target's passkey credential ID can impersonate that account without its authenticator, potentially gaining administrative control over SUPERADMIN accounts.

Zipline through 4.4.0 contains an authentication bypass vulnerability in the WebAuthn login endpoint that creates sessions for a supplied registered credential ID without verifying the assertion. Attackers who obtain a target's passkey credential ID can impersonate that account without its authenticator, potentially gaining administrative control over SUPERADMIN accounts.

Zipline before 4.5.1 contains a stored cross-site scripting vulnerability in the HighlightCode component that renders unescaped file content before highlight.js finishes loading. Authenticated attackers can upload text files containing HTML/JavaScript payloads that execute in viewers' browsers, enabling API calls such as account changes on the victim's behalf.

Zipline before 4.6.5 contains a TOCTOU race condition in the POST /api/setup endpoint that allows unauthenticated attackers to create multiple SUPERADMIN accounts by racing concurrent requests before the firstSetup flag is reset. Attackers can send simultaneous requests with spoofed X-Forwarded-For headers during the argon2 hashing window to gain full administrative control of the instance.

Zipline before 4.6.5 contains a TOCTOU race condition in the POST /api/setup endpoint that allows unauthenticated attackers to create multiple SUPERADMIN accounts by racing concurrent requests before the firstSetup flag is reset. Attackers can send simultaneous requests with spoofed X-Forwarded-For headers during the argon2 hashing window to gain full administrative control of the instance.

Zipline before 4.5.1 contains a stored cross-site scripting vulnerability in the HighlightCode component that renders unescaped file content before highlight.js finishes loading. Authenticated attackers can upload text files containing HTML/JavaScript payloads that execute in viewers' browsers, enabling API calls such as account changes on the victim's behalf.

OpenEMR before 8.4.0 contains OAuth scope canonicalization and containment flaws that allow unauthenticated attackers to obtain unapproved system-level FHIR access. Attackers can register a colon-form system:Patient.read client that bypasses approval, then request unregistered system/Patient.write scope to read, modify, and create Patient records.

Vvveb before 1.0.9.0 contains an authentication bypass vulnerability in generateCookie() in system/user/user.php that uses loose comparison to validate MD5 cookie hashes. Unauthenticated attackers can brute-force cookie values until the server's MD5 computation yields a 0e magic hash, then forge persistent cookies for account takeover.

Vvveb before 1.0.9.0 contains an authentication bypass vulnerability in generateCookie() in system/user/user.php that uses loose comparison to validate MD5 cookie hashes. Unauthenticated attackers can brute-force cookie values until the server's MD5 computation yields a 0e magic hash, then forge persistent cookies for account takeover.

Vvveb before 1.0.9.0 contains a denylist bypass in system/traits/media.php where '.phar' is misconfigured with a leading dot, allowing authenticated administrators to upload .phar PHP files. Attackers can upload a GIF magic-byte polyglot webshell bypassing binary inspection and execute arbitrary OS commands as the web-server user.