#git

Apple has publicly made available its Private Cloud Compute (PCC) Virtual Research Environment (VRE), allowing the research community to inspect and verify the privacy and security guarantees of its offering. PCC, which Apple unveiled earlier this June, has been marketed as the "most advanced security architecture ever deployed for cloud AI compute at scale." With the new technology, the idea is

Artificial Intelligence (AI) has rapidly evolved from a futuristic concept to a potent weapon in the hands of bad actors. Today, AI-based attacks are not just theoretical threats—they're happening across industries and outpacing traditional defense mechanisms.  The solution, however, is not futuristic. It turns out a properly designed identity security platform is able to deliver defenses

The Irish data protection watchdog on Thursday fined LinkedIn €310 million ($335 million) for violating the privacy of its users by conducting behavioral analyses of personal data for targeted advertising. "The inquiry examined LinkedIn's processing of personal data for the purposes of behavioral analysis and targeted advertising of users who have created LinkedIn profiles (members)," the Data

With the planned release of Red Hat Enterprise Linux (RHEL) 10 in 2025, the PKCS #12 (Public-Key Cryptography Standards #12) files created in FIPS mode now use Federal Information Processing Standard (FIPS) cryptography by default. In other words, PKCS #12 files allow for backup or easy transfer of keying material between RHEL systems using FIPS approved algorithms.What are PKCS #12 files?PKCS #12 is currently defined by RFC 7292 and is a format for storing and transferring private keys, certificates, and miscellaneous secrets. Typically, PKCS #12 is used for transferring private RSA, EdDSA, o

Vulnhuntr is a Python static code analyzer that uses Claude AI to find and explain complex, multistep vulnerabilities.

### Impact When a user disables two-factor authentication via the Panel, a `DELETE` request with their current password in a query parameter will be sent. While query parameters are encrypted when using TLS, many webservers (including ones officially documented for use with Pterodactyl) will log query parameters in plain-text, storing a user's password in plain text. If a malicious user obtains access to these logs they could *potentially* authenticate against a user's account; assuming they are able to discover the account's email address or username **separately**. ### Patches This problem has been patched by <https://github.com/pterodactyl/panel/commit/8be2b892c3940bdc0157ccdab16685a72d105dd1> on the `1.0-develop` branch and released under `v1.11.8` as a single commit on top of `v1.11.7` <https://github.com/pterodactyl/panel/commit/75b59080e2812ced677dab516222b2a3bb34e3a4> Patch file: <https://github.com/pterodactyl/panel/commit/8be2b892c3940bdc0157ccdab16685a72d105dd1.patch> ...

### Summary  The AWS Load Balancer Controller includes an optional, default-enabled feature that manages WAF WebACLs on Application Load Balancers (ALBs) on your behalf. In versions 2.8.1 and earlier, if the WebACL annotation [1] [alb.ingress.kubernetes.io/wafv2-acl-arn](http://alb.ingress.kubernetes.io/wafv2-acl-arn) or [alb.ingress.kubernetes.io/waf-acl-id](http://alb.ingress.kubernetes.io/waf-acl-id) was absent on Ingresses, the controller would automatically disassociate any existing WebACL from the ALBs, including those associated by AWS Firewall Manager (FMS). Customers on impacted versions should upgrade to prevent this issue from occurring. ### Impact  WebACLs attached to ALBs managed by the AWS Load Balancer Controller through methods other than Ingress annotations may be automatically removed, leaving the ALBs unprotected by WebACL.  Impacted versions: [>=2.0.0;<2.8.2]  ### Patches  We addressed this issue in version 2.8.2 [2] and recommend customers upgrade. Now, if t...

A government report's criticism of the 100x metric often used to justify fixing software earlier in development fuels a growing debate over pushing responsibility for secure code onto developers.

The load-language command expects a `lang` parameter from which it constructs the path of the localization file to load, of the form `translations-$LANG.json`. When doing so, it does not check that the resulting path is in the expected directory, which means that this command could be exploited to read other JSON files on the file system. The command should be patched by checking that the normalized path is in the expected directory.

php-heic-to-jpg <= 1.0.5 is vulnerable to remote code execution. An attacker who can upload heic images is able to execute code on the remote server via the file name. As a result, the CIA is no longer guaranteed. This affects php-heic-to-jpg 1.0.5 and below.