yiisoft/yii2

Yii 2 is a PHP application framework. Versions 2.0.54 and prior contain flawed logic in the core view rendering method View::renderPhpFile() that leads to Local File Inclusion. The function calls extract($_params_, EXTR_OVERWRITE) before the require statement that loads the view file. As a result, a caller-controlled _file_ key in the $params array overwrites the internal local variable specifying which file to include, potentially enabling RCE if an attacker can write PHP files through a separate primitive, as well as information disclosure. This issue has been fixed in version 2.0.55.

Yii 2 before 2.0.52 mishandles the attaching of behavior that is defined by an __class array key, a CVE-2024-4990 regression, as exploited in the wild in February through April 2025.

Yii 2 is a PHP application framework. During internal penetration testing of a product based on Yii2, users discovered a Cross-site Scripting (XSS) vulnerability within the framework itself. This issue is relevant for the latest version of Yii2 (2.0.49.3). This issue lies in the mechanism for displaying function argument values in the stack trace. The vulnerability manifests when an argument's value exceeds 32 characters. For convenience, argument values exceeding this limit are truncated and displayed with an added "...". The full argument value becomes visible when hovering over it with the mouse, as it is displayed in the title attribute of a span tag. However, the use of a double quote (") allows an attacker to break out of the title attribute's value context and inject their own attributes into the span tag, including malicious JavaScript code through event handlers such as onmousemove. This vulnerability allows an attacker to execute arbitrary JavaScript code in the security context of the victim's site via a specially crafted link. This could lead to the theft of cookies (including httpOnly cookies, which are accessible on the page), content substitution, or complete takeover of user accounts. This issue has been addressed in version 2.0.50. Users are advised to upgrade. There are no known workarounds for this vulnerability.

SQL injection vulnerability found in Yii Framework Yii 2 Framework before v.2.0.47 allows the a remote attacker to execute arbitrary code via the runAction function. NOTE: the software maintainer's position is that the vulnerability is in third-party code, not in the framework.

yii2 is vulnerable to Use of Predictable Algorithm in Random Number Generator

yii2 is vulnerable to Use of Predictable Algorithm in Random Number Generator

Yii 2 (yiisoft/yii2) before version 2.0.38 is vulnerable to remote code execution if the application calls `unserialize()` on arbitrary user input. This is fixed in version 2.0.38. A possible workaround without upgrading is available in the linked advisory.

Yii 2.x through 2.0.15.1 actively converts a wildcard CORS policy into reflecting an arbitrary Origin header value, which is incompatible with the CORS security design, and could lead to CORS misconfiguration security problems.

In Yii Framework 2.x before 2.0.14, remote attackers could obtain potentially sensitive information from exception messages, or exploit reflected XSS on the error handler page in non-debug mode. Related to base/ErrorHandler.php, log/Dispatcher.php, and views/errorHandler/exception.php.

In Yii Framework 2.x before 2.0.14, the switchIdentity function in web/User.php did not regenerate the CSRF token upon a change of identity.

An XSS vulnerability exists in framework/views/errorHandler/exception.php in Yii Framework 2.0.12 affecting the exception screen when debug mode is enabled, because $exception->errorInfo is mishandled.

Reflected Cross-site scripting (XSS) vulnerability in Yii Framework before 2.0.11, when development mode is used, allows remote attackers to inject arbitrary web script or HTML via crafted request data that is mishandled on the debug-mode exception screen.

Cross-site scripting (XSS) vulnerability in Yii Framework before 2.0.4 allows remote attackers to inject arbitrary web script or HTML via vectors related to JSON, arrays, and Internet Explorer 6 or 7.