#git

A stored cross-site scripting (XSS) vulnerability in the cms/content/edit component of YZNCMS v1.3.0 allows attackers to execute arbitrary web scripts or HTML via a crafted payload injected into the title parameter.

An issue in PGYER codefever v.2023.8.14-2ce4006 allows a remote attacker to execute arbitrary code via a crafted request to the branchList component.

SQL injection can exist in a newly created part of the JFinalcms background, and the parameters submitted by users are not filtered. As a result, special characters in parameters destroy the original logic of SQL statements. Attackers can use this vulnerability to execute any SQL statement.

### Impact This vulnerability affects deployments of Imageflow that involve decoding or processing malicious source .webp files. If you only process your own trusted files, this should not affect you (but you should update anyway). Imageflow relies on Google's [libwebp] library to decode .webp images, and is affected by the recent zero-day out-of-bounds write vulnerability [CVE-2023-4863](https://nvd.nist.gov/vuln/detail/CVE-2023-4863) and https://github.com/advisories/GHSA-j7hp-h8jx-5ppr. The libwebp vulnerability also affects Chrome, Android, macOS, and other consumers of the library). libwebp patched [the vulnerability](https://github.com/webmproject/libwebp/commit/2af26267cdfcb63a88e5c74a85927a12d6ca1d76 ) and released [1.3.2](https://github.com/webmproject/libwebp/releases/tag/v1.3.2) This was patched in [libwebp-sys in 0.9.3 and 0.9.4](https://github.com/NoXF/libwebp-sys/commits/master) **[Imageflow v2.0.0-preview8](https://github.com/imazen/imageflow/releases/tag/v2.0.0-p...

matrix-hookshot is a Matrix bot for connecting to external services like GitHub, GitLab, JIRA, and more. Instances that have enabled transformation functions (those that have `generic.allowJsTransformationFunctions` in their config), may be vulnerable to an attack where it is possible to break out of the `vm2` sandbox and as a result Hookshot will be vulnerable to this. This problem is only likely to affect users who have allowed untrusted users to apply their own transformation functions. If you have only enabled a limited set of trusted users, this threat is reduced (though not eliminated). Version 4.5.0 and above of hookshot include a new sandbox library which should better protect users. Users are advised to upgrade. Users unable to upgrade should disable `generic.allowJsTransformationFunctions` in the config.

Argo CD is a declarative continuous deployment framework for Kubernetes. In Argo CD versions prior to 2.3 (starting at least in v0.1.0, but likely in any version using Helm before 2.3), using a specifically-crafted Helm file could reference external Helm charts handled by the same repo-server to leak values, or files from the referenced Helm Chart. This was possible because Helm paths were predictable. The vulnerability worked by adding a Helm chart that referenced Helm resources from predictable paths. Because the paths of Helm charts were predictable and available on an instance of repo-server, it was possible to reference and then render the values and resources from other existing Helm charts regardless of permissions. While generally, secrets are not stored in these files, it was nevertheless possible to reference any values from these charts. This issue was fixed in Argo CD 2.3 and subsequent versions by randomizing Helm paths. User's still using Argo CD 2.3 or below are advised ...

### Impact Heap buffer overflow in `libwebp` allows a remote attacker to perform an out of bounds memory write via a crafted webp image. ### References - https://github.com/advisories/GHSA-j7hp-h8jx-5ppr - https://blog.isosceles.com/the-webp-0day/

The current regex implementation for parsing values in the module is susceptible to excessive backtracking, leading to potential DoS attacks. The regex implementation in question is as follows: ```js const functionNameMatch = /\s*function(?:\s|\s*\/\*[^(?:*/)]+\*\/\s*)*([^\s(/]+)/; ``` This vulnerability can be exploited when there is an imbalance in parentheses, which results in excessive backtracking and subsequently increases the CPU load and processing time significantly. This vulnerability can be triggered using the following input: ```js '\t'.repeat(54773) + '\t/function/i' ``` Here is a simple PoC code to demonstrate the issue: ```js const protocolre = /\sfunction(?:\s|\s/*[^(?:*\/)]+*/\s*)*([^\(\/]+)/; const startTime = Date.now(); const maliciousInput = '\t'.repeat(54773) + '\t/function/i' protocolre.test(maliciousInput); const endTime = Date.now(); console.log("process time: ", endTime - startTime, "ms"); ```

Sourcecodester Expense Tracker App v1 is vulnerable to Cross Site Scripting (XSS) via add category.

Sourcecodester Toll Tax Management System v1 is vulnerable to SQL Injection.