CVE-2026-97025

Flatpak writes the OCI repository authentication token with world-readable permissions 0644 in the system-helper's cache directory, allowing other local users on a multi-user system to read the token and impersonate the authenticated user against the OCI repository. Only OCI-based sources e.g. as...

CVE-2026-97025 Flatpak: flatpak: world-readable oci authentication token in system-helper cache path

Flatpak writes the OCI repository authentication token with world-readable permissions 0644 in the system-helper's cache directory, allowing other local users on a multi-user system to read the token and impersonate the authenticated user against the OCI repository. Only OCI-based sources e.g. as...

CVE-2026-97025

Flatpak writes the OCI repository authentication token with world-readable permissions 0644 in the system-helper's cache directory, allowing other local users on a multi-user system to read the token and impersonate the authenticated user against the OCI repository. Only OCI-based sources e.g. as...

EUVD-2026-88027

On a multi-user system, a user with an active local login session could downgrade a system-wide Flatpak app to an older version by removing the app's remote ref via the unprivileged system-helper RemoveLocalRef method, causing the anti-downgrade check to fail to find a reference date. A malicious...

CVE-2026-96281

On a multi-user system, a user with an active local login session could downgrade a system-wide Flatpak app to an older version by removing the app's remote ref via the unprivileged system-helper RemoveLocalRef method, causing the anti-downgrade check to fail to find a reference date. A malicious...

CVE-2026-96281

On a multi-user system, a user with an active local login session could downgrade a system-wide Flatpak app to an older version by removing the app's remote ref via the unprivileged system-helper RemoveLocalRef method, causing the anti-downgrade check to fail to find a reference date. A malicious...

CVE-2026-79721

MLflow platform (version 0.0.1 or newer ) is vulnerable to arbitrary code execution via a maliciously crafted model artifact . When an end user loads a tampered artifact into a project, it can execute attacker-controlled code on their system. The CNA-assigned CVSS 4.0 score is 8.6 (HIGH) , reflec...

CVE-2026-79721

Code execution can occur in versions of the MLflow platform running version 0.0.1 or newer, enabling a maliciously crafted model artifact to execute arbitrary code on an end user's system when loaded by the project...

EUVD-2026-74038

Code execution can occur in versions of the MLflow platform running version 0.0.1 or newer, enabling a maliciously crafted model artifact to execute arbitrary code on an end user's system when loaded by the project...

MSC-2026-11529

This package has been identified by Mend as containing potential malicious functionality. The severity of the functionality can change depending on where the library is running user's machine or backend server. The following risks were identified: • Contacting external host – this package sends...

MSC-2026-11517

This package has been identified by Mend as containing potential malicious functionality. The severity of the functionality can change depending on where the library is running user's machine or backend server. The following risks were identified: • Contacting external host – this package sends...

MSC-2026-11516

This package has been identified by Mend as containing potential malicious functionality. The severity of the functionality can change depending on where the library is running user's machine or backend server. The following risks were identified: • Contacting external host – this package sends...

MSC-2026-11493

This package has been identified by Mend as containing potential malicious functionality. The severity of the functionality can change depending on where the library is running user's machine or backend server. The following risks were identified: • Contacting external host – this package sends...

MSC-2026-11491

This package has been identified by Mend as containing potential malicious functionality. The severity of the functionality can change depending on where the library is running user's machine or backend server. The following risks were identified: • Contacting external host – this package sends...

MSC-2026-11486

This package has been identified by Mend as containing potential malicious functionality. The severity of the functionality can change depending on where the library is running user's machine or backend server. The following risks were identified: • Contacting external host – this package sends...

MSC-2026-11484

This package has been identified by Mend as containing potential malicious functionality. The severity of the functionality can change depending on where the library is running user's machine or backend server. The following risks were identified: • Contacting external host – this package sends...

MSC-2026-11482

This package has been identified by Mend as containing potential malicious functionality. The severity of the functionality can change depending on where the library is running user's machine or backend server. The following risks were identified: • Contacting external host – this package sends...

CVE-2026-6511

CVE-2026-6511 : Lenovo Smart Connect for Windows is reported to have an improper access control vulnerability that could allow a local authenticated user to access files owned by another user on the same system. The issue is described as local, with low attack complexity and the attack requires l...

llama-index-core insecurely handles temporary files

The llama-index-core package, up to version 0.12.44, contains a vulnerability in the getcachedir function where a predictable, hardcoded directory path /tmp/llamaindex is used on Linux systems without proper security controls. This vulnerability allows attackers on multi-user systems to steal...

PYSEC-2026-1562 llama-index-core insecurely handles temporary files

The llama-index-core package, up to version 0.12.44, contains a vulnerability in the getcachedir function where a predictable, hardcoded directory path /tmp/llamaindex is used on Linux systems without proper security controls. This vulnerability allows attackers on multi-user systems to steal...