#vulnerability

Prompt injection attacks could be coming to an AI browser near you. Read on to understand what these attacks do and how to stay safe.

## **Vulnerability Details** The magnified size calculations in `ReadOneMNGIMage` (in `coders/png.c`) are unsafe and can overflow, leading to memory corruption. The source snippet below is heavily abbreviated due to the size of the function, but hopefully the important points are captured. ```c static Image *ReadOneMNGImage(MngReadInfo* mng_info, const ImageInfo *image_info,ExceptionInfo *exception) { // Lots of stuff, this is effectively a state machine for the MNG rendering commands, // skip to the point where we start processing the "MAGN" command. if (memcmp(type,mng_MAGN,4) == 0) { png_uint_16 magn_first, magn_last, magn_mb, magn_ml, magn_mr, magn_mt, magn_mx, magn_my, magn_methx, magn_methy; // Details unimportant, but each of the `magn_xxx` variables is read from the file. if (magn_first == ...

This vulnerability was discovered by researchers at **Check Point**. We are sharing this report as part of a responsible disclosure process and are happy to assist in validation and remediation if needed. ### Summary A privilege escalation vulnerability exists in Langflow containers where an authenticated user with RCE access can invoke the internal CLI command **langflow superuser** to create a new administrative user. This results in full superuser access, even if the user initially registered through the UI as a regular (non-admin) account. ### Details Langflow's Docker image includes a CLI binary at /app/.venv/bin/langflow that exposes sensitive commands, including: `langflow superuser` This command allows creation of a new superuser without checking whether one already exists. When combined with code execution (e.g., via the authenticated **/api/v1/validate/code** endpoint), a low-privileged user can execute: `/app/.venv/bin/langflow superuser` inside the container, and el...

## **Vulnerability Details** When performing image magnification in `ReadOneMNGIMage` (in `coders/png.c`), there is an issue around the handling of images with separate alpha channels. When loading an image with a color type that implies a separate alpha channel (ie. `jng_color_type >= 12`), we will load the alpha pixels in this loop: ```c if (logging != MagickFalse) (void) LogMagickEvent(CoderEvent,GetMagickModule(), " Reading alpha from alpha_blob."); jng_image=ReadImage(alpha_image_info,exception); if (jng_image != (Image *) NULL) for (y=0; y < (ssize_t) image->rows; y++) { s=GetVirtualPixels(jng_image,0,y,image->columns,1,exception); q=GetAuthenticPixels(image,0,y,image->columns,1,exception); // [0] if ((s == (const Quantum *) NULL) || (q == (Quantum *) NULL)) break; if (image->alpha_trait != UndefinedPixelTrait) for (x=(ssize_t) image->columns; x != 0; x--) ...

## Summary In ImageMagick's `magick stream` command, specifying multiple consecutive `%d` format specifiers in a filename template causes a memory leak. ## Details - **Vulnerability Type:** Memory leak - **Affected Version:** ImageMagick 7.1.1-47 (as of commit 82572afc, June 2025) ## Reproduction ### Tested Environment - **Operating System:** Ubuntu 22.04 LTS - **Architecture:** x86_64 - **Compiler:** gcc with AddressSanitizer (gcc version: 11.4.0) ### Reproduction Steps ```bash # Clone source git clone --depth 1 --branch 7.1.1-47 https://github.com/ImageMagick/ImageMagick.git ImageMagick-7.1.1 cd ImageMagick-7.1.1 # Build with ASan CFLAGS="-g -O0 -fsanitize=address -fno-omit-frame-pointer" CXXFLAGS="$CFLAGS" LDFLAGS="-fsanitize=address" ./configure --enable-maintainer-mode --enable-shared && make -j$(nproc) && make install # Trigger crash ./utilities/magick stream %d%d a a ``` ### Output ``` $ magick stream %d%d a a stream: no decode delegate for this image format `' @ erro...

## Summary In ImageMagick's `magick mogrify` command, specifying multiple consecutive `%d` format specifiers in a filename template causes internal pointer arithmetic to generate an address below the beginning of the stack buffer, resulting in a stack overflow through `vsnprintf()`. ## Details - **Vulnerability Type:** CWE-124: Buffer Underwrite - **Affected Component:** MagickCore/image.c - Format processing within InterpretImageFilename() - **Affected Version:** ImageMagick 7.1.1-47 (as of commit 82572afc, June 2025) - **CWE-124: Buffer Underwrite:** A vulnerability where writing occurs to memory addresses before the beginning of a buffer. This is caused by a design flaw in fixed offset correction, resulting in negative pointer arithmetic during consecutive format specifier processing. ## Reproduction ### Tested Environment - **Operating System:** Ubuntu 22.04 LTS - **Architecture:** x86_64 - **Compiler:** gcc with AddressSanitizer (gcc version: 11.4.0) ### Reproduction Steps ...

Privilege Defined With Unsafe Actions vulnerability in Apache Cassandra. An user with MODIFY permission ON ALL KEYSPACES can escalate privileges to superuser within a targeted Cassandra cluster via unsafe actions to a system resource. Operators granting data MODIFY permission on all keyspaces on affected versions should review data access rules for potential breaches. This issue affects Apache Cassandra 3.0.30, 3.11.17, 4.0.16, 4.1.7, 5.0.2, but this advisory is only for 4.0.16 because the fix to CVE-2025-23015 was incorrectly applied to 4.0.16, so that version is still affected. Users in the 4.0 series are recommended to upgrade to version 4.0.17 which fixes the issue. Users from 3.0, 3.11, 4.1 and 5.0 series should follow recommendation from CVE-2025-23015.

**Product:** PhpSpreadsheet **Version:** 3.8.0 **CWE-ID:** CWE-918: Server-Side Request Forgery (SSRF) **CVSS vector v.3.1:** 7.5 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N) **CVSS vector v.4.0:** 8.7 (AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N) **Description:** SSRF occurs when a processed HTML document is read and displayed in the browser **Impact:** Server-Side Request Forgery **Vulnerable component:** the `PhpOffice\PhpSpreadsheet\Worksheet\Drawing` class, `setPath` method **Exploitation conditions:** getting a string from the user that is passed to the HTML reader **Mitigation:** improved processing of the `$path` variable of the `setPath` method of the `PhpOffice\PhpSpreadsheet\Worksheet\Drawing` class is needed **Researcher: Aleksey Solovev (Positive Technologies)** # Research The researcher discovered zero-day vulnerability Server-Side Request Forgery (SSRF) (in the `setPath` method of the `PhpOffice\PhpSpreadsheet\Worksheet\Drawing` class) in Phpspreadsheet. The latest ...

A list of topics we covered in the week of August 18 to August 24 of 2025

At Microsoft, securing the ecosystem means more than just fixing bugs—it means proactively hunting for variant classes, identifying systemic weaknesses, and working across teams to protect customers before attackers ever get the chance. This blog highlights one such effort: a deep dive into the risks of misconfigured postMessage handlers across Microsoft services and how MSRC worked with engineering teams to mitigate them.