🚨 CVE-2026-46432
LMDeploy is a toolkit for compressing, deploying, and serving large language models. In versions 0.12.3 and prior, LMDeploy is vulnerable to arbitrary code execution through hardcoded "trust_remote_code=True" in multiple HuggingFace model-loading call sites. At time of publication, there are no publicly available patches.
🎖@cveNotify
LMDeploy is a toolkit for compressing, deploying, and serving large language models. In versions 0.12.3 and prior, LMDeploy is vulnerable to arbitrary code execution through hardcoded "trust_remote_code=True" in multiple HuggingFace model-loading call sites. At time of publication, there are no publicly available patches.
🎖@cveNotify
GitHub
Arbitrary code execution via hardcoded trust_remote_code=True in lmdeploy model initialization
## Summary
lmdeploy hardcodes `trust_remote_code=True` in multiple HuggingFace model-loading call sites.
The affected code paths are in:
```text
lmdeploy/archs.py
lmdeploy/utils.py
````...
lmdeploy hardcodes `trust_remote_code=True` in multiple HuggingFace model-loading call sites.
The affected code paths are in:
```text
lmdeploy/archs.py
lmdeploy/utils.py
````...
🚨 CVE-2026-46518
OpenEMR is a free and open source electronic health records and medical practice management application. Prior to version 8.0.0.1, a stored cross-site scripting vulnerability in the prescription CSS/HTML multi-print feature allows a patient portal user to execute arbitrary JavaScript in a clinician's browser session. Patient demographic fields (name, address) are rendered without output encoding in multiprintcss_header(), and portal patients can write attacker-controlled HTML directly into patient_data by calling the PUT api/patient/:num endpoint, which bypasses the intended audit review workflow. Because the XSS fires in the clinician's authenticated session on the main OpenEMR interface, the attacker can access CSRF tokens, session data, and perform actions as the clinician — crossing the patient-to-clinician trust boundary. This issue has been patched in version 8.0.0.1.
🎖@cveNotify
OpenEMR is a free and open source electronic health records and medical practice management application. Prior to version 8.0.0.1, a stored cross-site scripting vulnerability in the prescription CSS/HTML multi-print feature allows a patient portal user to execute arbitrary JavaScript in a clinician's browser session. Patient demographic fields (name, address) are rendered without output encoding in multiprintcss_header(), and portal patients can write attacker-controlled HTML directly into patient_data by calling the PUT api/patient/:num endpoint, which bypasses the intended audit review workflow. Because the XSS fires in the clinician's authenticated session on the main OpenEMR interface, the attacker can access CSRF tokens, session data, and perform actions as the clinician — crossing the patient-to-clinician trust boundary. This issue has been patched in version 8.0.0.1.
🎖@cveNotify
GitHub
vuln06: Stored XSS in prescription CSS/HTML print view via patient demographics
### Summary
A stored cross-site scripting vulnerability in the prescription CSS/HTML multi-print feature allows a patient portal user to execute arbitrary JavaScript in a clinician's browser...
A stored cross-site scripting vulnerability in the prescription CSS/HTML multi-print feature allows a patient portal user to execute arbitrary JavaScript in a clinician's browser...
🚨 CVE-2024-58350
Ghidra before 11.2 contains a use after free vulnerability in the Sleigh backend caused by undefined static initialization order of the SleighArchitecture::translators and XmlArchitectureCapability singletons. Attackers can trigger an infinite loop or denial of service during shutdown by exploiting the unsafe destruction order that causes iteration over deallocated memory.
🎖@cveNotify
Ghidra before 11.2 contains a use after free vulnerability in the Sleigh backend caused by undefined static initialization order of the SleighArchitecture::translators and XmlArchitectureCapability singletons. Attackers can trigger an infinite loop or denial of service during shutdown by exploiting the unsafe destruction order that causes iteration over deallocated memory.
🎖@cveNotify
GitHub
Use after free vulnerability in Sleigh backend
### Summary
This is an SIOF issue (ref: https://en.cppreference.com/w/cpp/language/siof). There are two globals in different translation units, and therefore their destruction order is undefined,...
This is an SIOF issue (ref: https://en.cppreference.com/w/cpp/language/siof). There are two globals in different translation units, and therefore their destruction order is undefined,...
🚨 CVE-2025-71329
image-size through 2.0.2 contains a denial of service vulnerability that allows remote attackers to permanently block the Node.js event loop by supplying a specially crafted image buffer with a zero-valued size field in a recognized box-type. Attackers can trigger an infinite loop in the JXL or HEIF image parsers by providing a crafted image containing a box with a size of zero, causing the offset to never advance and permanently hanging the application.
🎖@cveNotify
image-size through 2.0.2 contains a denial of service vulnerability that allows remote attackers to permanently block the Node.js event loop by supplying a specially crafted image buffer with a zero-valued size field in a recognized box-type. Attackers can trigger an infinite loop in the JXL or HEIF image parsers by providing a crafted image containing a box with a size of zero, causing the offset to never advance and permanently hanging the application.
🎖@cveNotify
Joshua Rogers’ Scribbles
Two infinite loop / DoS vulnerabilities in image-size
Two infinite loop / Denial of Service vulnerabilities I found while auditing the npm package image-size, affecting its HEIF, JP2, JXL, and ICNS parsing in every version up to at least 2.0.2.
🚨 CVE-2025-71330
image-size through 2.0.2 contains a denial of service vulnerability that allows remote attackers to permanently block the Node.js event loop by supplying a specially crafted ICNS image buffer. Attackers can craft an ICNS buffer containing valid magic bytes and a zero-valued entry length field to trigger an infinite loop in the ICNS parser, as the offset is never incremented when the entry length field is 0, causing the while loop condition to remain true indefinitely.
🎖@cveNotify
image-size through 2.0.2 contains a denial of service vulnerability that allows remote attackers to permanently block the Node.js event loop by supplying a specially crafted ICNS image buffer. Attackers can craft an ICNS buffer containing valid magic bytes and a zero-valued entry length field to trigger an infinite loop in the ICNS parser, as the offset is never incremented when the entry length field is 0, causing the while loop condition to remain true indefinitely.
🎖@cveNotify
Joshua Rogers’ Scribbles
Two infinite loop / DoS vulnerabilities in image-size
Two infinite loop / Denial of Service vulnerabilities I found while auditing the npm package image-size, affecting its HEIF, JP2, JXL, and ICNS parsing in every version up to at least 2.0.2.
🚨 CVE-2026-49069
Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') vulnerability in WPZOOM Portfolio allows Reflected XSS.
This issue affects WPZOOM Portfolio: from n/a through 1.4.21.
🎖@cveNotify
Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') vulnerability in WPZOOM Portfolio allows Reflected XSS.
This issue affects WPZOOM Portfolio: from n/a through 1.4.21.
🎖@cveNotify
Patchstack
Cross Site Scripting (XSS) in WordPress WPZOOM Portfolio Plugin
Patchstack is the leading open source vulnerability research organization. Find information and protection for all WordPress, Drupal and Joomla security issues.
🚨 CVE-2026-49495
Ghidra 10.2 before 12.1 contains an uncontrolled resource consumption vulnerability in ExportTrie.parseTrie() that lacks cycle detection when traversing Mach-O binary export tries. A crafted Mach-O binary with circular references in the export trie causes unbounded queue growth and exponential string concatenation, triggering OutOfMemoryError that crashes the entire JVM and loses all unsaved work.
🎖@cveNotify
Ghidra 10.2 before 12.1 contains an uncontrolled resource consumption vulnerability in ExportTrie.parseTrie() that lacks cycle detection when traversing Mach-O binary export tries. A crafted Mach-O binary with circular references in the export trie causes unbounded queue growth and exponential string concatenation, triggering OutOfMemoryError that crashes the entire JVM and loses all unsaved work.
🎖@cveNotify
GitHub
Uncontrolled Resource Consumption via Circular Reference in Mach-O Export Trie Parser
## Summary
`ExportTrie.parseTrie()` traverses a Mach-O binary's export trie using BFS with no cycle detection. A crafted trie containing a circular reference causes unbounded queue growth an...
`ExportTrie.parseTrie()` traverses a Mach-O binary's export trie using BFS with no cycle detection. A crafted trie containing a circular reference causes unbounded queue growth an...
🚨 CVE-2026-10879
DBI versions before 1.648 for Perl have a heap overflow when preparsing SQL statements with more than 9 binders.
The preparse method expands SQL placeholder characters to numbered binders of the form :pN, but only allocates three characters per binder in the buffer. Placeholders 10-99 require four characters, 100-999 require five characters, et cetera.
🎖@cveNotify
DBI versions before 1.648 for Perl have a heap overflow when preparsing SQL statements with more than 9 binders.
The preparse method expands SQL placeholder characters to numbered binders of the form :pN, but only allocates three characters per binder in the buffer. Placeholders 10-99 require four characters, 100-999 require five characters, et cetera.
🎖@cveNotify
🚨 CVE-2026-11362
DataDog::DogStatsd versions through 0.07 for Perl allow metric injections from event tags.
DataDog::DogStatsd does not properly sanitise input, allowing metric injections of data from untrusted sources.
The format_event method (used by the event method) does not validate the content of the tags, which may contain commas (allowing tags to be injected) or newlines, pipes and colons that allow metric injections. (There is an ineffective s/|//g to remove pipes, but because the pipe is not escaped, it is interpreted as a regular expression metacharacter and has no effect.)
🎖@cveNotify
DataDog::DogStatsd versions through 0.07 for Perl allow metric injections from event tags.
DataDog::DogStatsd does not properly sanitise input, allowing metric injections of data from untrusted sources.
The format_event method (used by the event method) does not validate the content of the tags, which may contain commas (allowing tags to be injected) or newlines, pipes and colons that allow metric injections. (There is an ineffective s/|//g to remove pipes, but because the pipe is not escaped, it is interpreted as a regular expression metacharacter and has no effect.)
🎖@cveNotify
🚨 CVE-2026-9270
DataDog::DogStatsd versions through 0.07 for Perl allow metric injections.
DataDog::DogStatsd does not properly sanitise input, allowing metric injections of data from untrusted sources.
The send_stats method does not remove newlines from metric names ($stat variable), allowing attackers to change the metric name prefix.
The send_stats method does not validate the content of the value ($delta variable), allowing attackers to inject metrics, especially from methods that do not restrict the data type for the value, such as set, gauge, count and histogram.
The send_stats method does not validate the content of the tags, which may contain newlines, pipes and colons that allow metric injections.
Note that the SYNOPSIS shows an example of passing a website form "loginName" parameter as a tag, which is unsafe.
🎖@cveNotify
DataDog::DogStatsd versions through 0.07 for Perl allow metric injections.
DataDog::DogStatsd does not properly sanitise input, allowing metric injections of data from untrusted sources.
The send_stats method does not remove newlines from metric names ($stat variable), allowing attackers to change the metric name prefix.
The send_stats method does not validate the content of the value ($delta variable), allowing attackers to inject metrics, especially from methods that do not restrict the data type for the value, such as set, gauge, count and histogram.
The send_stats method does not validate the content of the tags, which may contain newlines, pipes and colons that allow metric injections.
Note that the SYNOPSIS shows an example of passing a website form "loginName" parameter as a tag, which is unsafe.
🎖@cveNotify
🚨 CVE-2026-10725
Protocol::HTTP2 versions before 1.13 for Perl is vulnerable to a HTTP/2 Bomb.
Protocol::HTTP2's inbound HPACK path has no header-list size limit, so a small HTTP/2 request can expand into large server memory (the "HTTP/2 bomb").
The headers_decode method materialises a full key+value copy per indexed reference with no running size check, and the stream_header_block_add method appends (since version 1.12) every CONTINUATION frame to the per-stream buffer unbounded.
MAX_HEADER_LIST_SIZE (default 65536) is advertised in SETTINGS but never consulted on decode. It is absent from the decoder and from the :limits export tag.
🎖@cveNotify
Protocol::HTTP2 versions before 1.13 for Perl is vulnerable to a HTTP/2 Bomb.
Protocol::HTTP2's inbound HPACK path has no header-list size limit, so a small HTTP/2 request can expand into large server memory (the "HTTP/2 bomb").
The headers_decode method materialises a full key+value copy per indexed reference with no running size check, and the stream_header_block_add method appends (since version 1.12) every CONTINUATION frame to the per-stream buffer unbounded.
MAX_HEADER_LIST_SIZE (default 65536) is advertised in SETTINGS but never consulted on decode. It is absent from the decoder and from the :limits export tag.
🎖@cveNotify
🚨 CVE-2023-42456
Sudo-rs, a memory safe implementation of sudo and su, allows users to not have to enter authentication at every sudo attempt, but instead only requiring authentication every once in a while in every terminal or process group. Only once a configurable timeout has passed will the user have to re-authenticate themselves. Supporting this functionality is a set of session files (timestamps) for each user, stored in `/var/run/sudo-rs/ts`. These files are named according to the username from which the sudo attempt is made (the origin user).
An issue was discovered in versions prior to 0.2.1 where usernames containing the `.` and `/` characters could result in the corruption of specific files on the filesystem. As usernames are generally not limited by the characters they can contain, a username appearing to be a relative path can be constructed. For example we could add a user to the system containing the username `../../../../bin/cp`. When logged in as a user with that name, that user could run `sudo -K` to clear their session record file. The session code then constructs the path to the session file by concatenating the username to the session file storage directory, resulting in a resolved path of `/bin/cp`. The code then clears that file, resulting in the `cp` binary effectively being removed from the system.
An attacker needs to be able to login as a user with a constructed username. Given that such a username is unlikely to exist on an existing system, they will also need to be able to create the users with the constructed usernames.
The issue is patched in version 0.2.1 of sudo-rs. Sudo-rs now uses the uid for the user instead of their username for determining the filename. Note that an upgrade to this version will result in existing session files being ignored and users will be forced to re-authenticate. It also fully eliminates any possibility of path traversal, given that uids are always integer values.
The `sudo -K` and `sudo -k` commands can run, even if a user has no sudo access. As a workaround, make sure that one's system does not contain any users with a specially crafted username. While this is the case and while untrusted users do not have the ability to create arbitrary users on the system, one should not be able to exploit this issue.
🎖@cveNotify
Sudo-rs, a memory safe implementation of sudo and su, allows users to not have to enter authentication at every sudo attempt, but instead only requiring authentication every once in a while in every terminal or process group. Only once a configurable timeout has passed will the user have to re-authenticate themselves. Supporting this functionality is a set of session files (timestamps) for each user, stored in `/var/run/sudo-rs/ts`. These files are named according to the username from which the sudo attempt is made (the origin user).
An issue was discovered in versions prior to 0.2.1 where usernames containing the `.` and `/` characters could result in the corruption of specific files on the filesystem. As usernames are generally not limited by the characters they can contain, a username appearing to be a relative path can be constructed. For example we could add a user to the system containing the username `../../../../bin/cp`. When logged in as a user with that name, that user could run `sudo -K` to clear their session record file. The session code then constructs the path to the session file by concatenating the username to the session file storage directory, resulting in a resolved path of `/bin/cp`. The code then clears that file, resulting in the `cp` binary effectively being removed from the system.
An attacker needs to be able to login as a user with a constructed username. Given that such a username is unlikely to exist on an existing system, they will also need to be able to create the users with the constructed usernames.
The issue is patched in version 0.2.1 of sudo-rs. Sudo-rs now uses the uid for the user instead of their username for determining the filename. Note that an upgrade to this version will result in existing session files being ignored and users will be forced to re-authenticate. It also fully eliminates any possibility of path traversal, given that uids are always integer values.
The `sudo -K` and `sudo -k` commands can run, even if a user has no sudo access. As a workaround, make sure that one's system does not contain any users with a specially crafted username. While this is the case and while untrusted users do not have the ability to create arbitrary users on the system, one should not be able to exploit this issue.
🎖@cveNotify
GitHub
Merge pull request from GHSA-2r3c-m6v7-9354 · trifectatechfoundation/sudo-rs@bfdbda2
Use uid instead of username for storing session files
🚨 CVE-2026-11029
Insufficient validation of untrusted input in Drag and Drop in Google Chrome on Android prior to 149.0.7827.53 allowed a remote attacker who had compromised the renderer process to potentially perform a sandbox escape via a crafted HTML page. (Chromium security severity: Medium)
🎖@cveNotify
Insufficient validation of untrusted input in Drag and Drop in Google Chrome on Android prior to 149.0.7827.53 allowed a remote attacker who had compromised the renderer process to potentially perform a sandbox escape via a crafted HTML page. (Chromium security severity: Medium)
🎖@cveNotify
Chrome Releases
Stable Channel Update for Desktop
The Chrome team is delighted to announce the promotion of Chrome 149 to the stable channel for Windows, Mac and Linux. This will roll out ov...
🚨 CVE-2026-11676
Insufficient validation of untrusted input in Dawn in Google Chrome on Linux and ChromeOS prior to 149.0.7827.103 allowed a remote attacker who had compromised the renderer process to potentially perform a sandbox escape via a crafted HTML page. (Chromium security severity: High)
🎖@cveNotify
Insufficient validation of untrusted input in Dawn in Google Chrome on Linux and ChromeOS prior to 149.0.7827.103 allowed a remote attacker who had compromised the renderer process to potentially perform a sandbox escape via a crafted HTML page. (Chromium security severity: High)
🎖@cveNotify
Chrome Releases
Stable Channel Update for Desktop
The Stable channel has been updated to 149.0.7827.102/.103 for Windows and Mac and 149.0.7827.102 for Linux, which will roll out over the ...
🚨 CVE-2026-11681
Use after free in Ozone in Google Chrome on Linux prior to 149.0.7827.103 allowed a remote attacker to potentially exploit heap corruption via a crafted HTML page. (Chromium security severity: High)
🎖@cveNotify
Use after free in Ozone in Google Chrome on Linux prior to 149.0.7827.103 allowed a remote attacker to potentially exploit heap corruption via a crafted HTML page. (Chromium security severity: High)
🎖@cveNotify
Chrome Releases
Stable Channel Update for Desktop
The Stable channel has been updated to 149.0.7827.102/.103 for Windows and Mac and 149.0.7827.102 for Linux, which will roll out over the ...
🚨 CVE-2026-11682
Inappropriate implementation in Views in Google Chrome on Linux prior to 149.0.7827.103 allowed a remote attacker who had compromised the renderer process to potentially perform a sandbox escape via a crafted HTML page. (Chromium security severity: High)
🎖@cveNotify
Inappropriate implementation in Views in Google Chrome on Linux prior to 149.0.7827.103 allowed a remote attacker who had compromised the renderer process to potentially perform a sandbox escape via a crafted HTML page. (Chromium security severity: High)
🎖@cveNotify
Chrome Releases
Stable Channel Update for Desktop
The Stable channel has been updated to 149.0.7827.102/.103 for Windows and Mac and 149.0.7827.102 for Linux, which will roll out over the ...
🚨 CVE-2026-24064
Waves Central for macOS versions 13.0.9 through 16.5.5 contain a local privilege escalation vulnerability. A trusted XPC client component included with the product is signed with hardened runtime entitlements that permit dynamic library injection. A local attacker can set the DYLD_INSERT_LIBRARIES environment variable to inject an attacker-controlled dynamic library into the trusted client process at launch. The injected code runs within the signed process and can connect to the product's privileged helper service to invoke privileged operations, resulting in arbitrary code execution as root. The issue is fixed in version 16.6.2.
🎖@cveNotify
Waves Central for macOS versions 13.0.9 through 16.5.5 contain a local privilege escalation vulnerability. A trusted XPC client component included with the product is signed with hardened runtime entitlements that permit dynamic library injection. A local attacker can set the DYLD_INSERT_LIBRARIES environment variable to inject an attacker-controlled dynamic library into the trusted client process at launch. The injected code runs within the signed process and can connect to the product's privileged helper service to invoke privileged operations, resulting in arbitrary code execution as root. The issue is fixed in version 16.6.2.
🎖@cveNotify
🚨 CVE-2026-38615
DedeCMS V5.7.118 is vulnerable to Command Execution in file_manage_control.php.
🎖@cveNotify
DedeCMS V5.7.118 is vulnerable to Command Execution in file_manage_control.php.
🎖@cveNotify
GitHub
CMS/1.docx at main · Bul11et/CMS
Contribute to Bul11et/CMS development by creating an account on GitHub.
🚨 CVE-2026-45447
Issue summary: A specially crafted PKCS#7 or S/MIME signed message could
trigger a use-after-free during PKCS#7 signature verification.
Impact summary: A use-after-free may result in process crashes, heap
corruption, or potentially remote code execution.
When processing a PKCS#7 or S/MIME signed message, if the SignedData
digestAlgorithms field is present as an empty ASN.1 SET, OpenSSL may
incorrectly free a caller-owned BIO during PKCS7_verify(). A subsequent
use of the BIO by the calling application results in a use-after-free
condition.
In the common case this occurs when the application later calls
BIO_free() on the BIO originally passed to PKCS7_verify(). Depending
on allocator behavior and application-specific BIO usage patterns, this
may result in a crash or other memory corruption. In some application
contexts this may potentially be exploitable for remote code execution.
Applications that process PKCS#7 or S/MIME signed messages using OpenSSL
PKCS#7 APIs may be affected. Applications using the CMS APIs for this
processing are not affected.
The FIPS modules in 4.0, 3.6, 3.5, 3.4, and 3.0 are not affected by this
issue, as the affected code is outside the OpenSSL FIPS module boundary.
🎖@cveNotify
Issue summary: A specially crafted PKCS#7 or S/MIME signed message could
trigger a use-after-free during PKCS#7 signature verification.
Impact summary: A use-after-free may result in process crashes, heap
corruption, or potentially remote code execution.
When processing a PKCS#7 or S/MIME signed message, if the SignedData
digestAlgorithms field is present as an empty ASN.1 SET, OpenSSL may
incorrectly free a caller-owned BIO during PKCS7_verify(). A subsequent
use of the BIO by the calling application results in a use-after-free
condition.
In the common case this occurs when the application later calls
BIO_free() on the BIO originally passed to PKCS7_verify(). Depending
on allocator behavior and application-specific BIO usage patterns, this
may result in a crash or other memory corruption. In some application
contexts this may potentially be exploitable for remote code execution.
Applications that process PKCS#7 or S/MIME signed messages using OpenSSL
PKCS#7 APIs may be affected. Applications using the CMS APIs for this
processing are not affected.
The FIPS modules in 4.0, 3.6, 3.5, 3.4, and 3.0 are not affected by this
issue, as the affected code is outside the OpenSSL FIPS module boundary.
🎖@cveNotify
GitHub
Fix possible use-after-free in OpenSSL PKCS7_verify() · openssl/openssl@3aad5eb
Fixes CVE-2026-45447
Reviewed-by: Eugene Syromiatnikov <esyr@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.foundation>
MergeDate: Mon Jun 8 20:22:50 2026
(cherry...
Reviewed-by: Eugene Syromiatnikov <esyr@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.foundation>
MergeDate: Mon Jun 8 20:22:50 2026
(cherry...
🚨 CVE-2026-47641
Improper neutralization of input during web page generation ('cross-site scripting') in Microsoft Office SharePoint allows an authorized attacker to perform spoofing over a network.
🎖@cveNotify
Improper neutralization of input during web page generation ('cross-site scripting') in Microsoft Office SharePoint allows an authorized attacker to perform spoofing over a network.
🎖@cveNotify
🚨 CVE-2026-47935
Adobe Experience Manager versions 6.5.24, LTS SP1, 2026.04 and earlier are affected by a DOM-based Cross-Site Scripting (XSS) vulnerability. An attacker could exploit this issue by manipulating the DOM environment to execute malicious JavaScript within the context of the victim's browser. Exploitation of this issue requires user interaction in that a victim must visit a crafted webpage. Scope is changed.
🎖@cveNotify
Adobe Experience Manager versions 6.5.24, LTS SP1, 2026.04 and earlier are affected by a DOM-based Cross-Site Scripting (XSS) vulnerability. An attacker could exploit this issue by manipulating the DOM environment to execute malicious JavaScript within the context of the victim's browser. Exploitation of this issue requires user interaction in that a victim must visit a crafted webpage. Scope is changed.
🎖@cveNotify
Adobe
Adobe Security Bulletin
Security updates available for Adobe Experience Manager | APSB26-24