#open-source_1
-
Rust Adopts LLM Usage Rules for rust-lang/rust Contributors
Five teams within the Rust project have adopted a policy governing the use of large language models when working on the main repository. The core principle: AI may be used to analyze and verify, but not to create code.
-
Munich Will Fund Six Months of Work on libexpat
Under its Open Source Sabbatical program, the city of Munich has hired the maintainer of the popular XML parser libexpat: up to six months of paid work on the library's security and reliability.
-
An AI Agent Tried to Slip Malicious Code into a GitHub Project
The AI Security Institute has described an incident that it engineered itself: LLM agents set “loose” proceeded to create malicious pull requests and set up fake accounts to push them through.
-
Mistral Releases Shieldstral: An Open-Weight 3B Content Moderation Model
Mistral has unveiled Shieldstral, a compact safety classifier that takes moderation policies in natural language at runtime and, according to its developers, holds its own against models seven times larger.
-
The Linux Staging Subsystem Will Reject LLM Patches — Except for Genuine Security Fixes
Greg Kroah-Hartman has announced that the kernel's drivers/staging/ subtree will no longer accept patches generated by language models. The one exception: truly valid vulnerability fixes that have been verified on real hardware.
-
FFmpeg 9.0: More Acceleration via Vulkan, Animated WebP and New AMD AMF Filters
A new feature release of the popular open-source multimedia library FFmpeg is out. Version 9.0 expands hardware acceleration through Vulkan, adds support for animated WebP and a host of new filters.
-
C-Kermit 11: The First Release of a Legendary Utility in 15 Years
John Gurzen has released C-Kermit 11 — the first version in a decade and a half of what was once an indispensable tool for transferring files between machines.
-
Rust wants to get rid of Pin: immovable types and guaranteed destructors
The Rust team has adopted a project goal for 2026–2027: to introduce the Move and Forget traits, which would let types opt out of being moved and “forgotten.” This should simplify working with self-referential types and, eventually, fully supplant the Pin mechanism.
-
9front ships its “THIS WAS SUPPOSED TO BE FUN” release
The fork of Bell Labs' Plan 9 operating system gets a new version with a Synaptics touchpad driver, SHA3 support, x2APIC and a host of fixes to the gefs file system.
-
Shitty: a terminal with a dumb name and serious ambitions
The author, pg83, has unveiled Shitty — a terminal emulator that, judging by the benchmarks, outpaces alacritty, kitty and ghostty, while openly admitting to being “memory-unsafe.”
-
NetBSD 11.0 Released: RISC-V Support and Improved Linux Compatibility
The nineteenth major release of the BSD system adds support for the 64-bit RISC-V architecture, Qualcomm Snapdragon X1 Elite laptop chips, and expanded compatibility with Linux system calls.
-
GNOME Introduces an RFC Process and Battles a Flood of AI Extensions
The GNOME project closed out July working on a formal RFC process to govern its development while simultaneously fighting off a surge of sloppy, neural-network-generated GNOME Shell extensions.
-
Arch Linux Veteran Morten Linderud Leaves the Project After 10 Years
Developer, security team member and maintainer of numerous packages Morten “Foxboron” Linderud has announced his departure from Arch Linux, leaving a number of key packages without a maintainer.
-
84 Vulnerabilities in 4G and 5G Network Cores: From Denial of Service to Session Hijacking
Researchers from Nanyang Technological University have uncovered an entire class of vulnerabilities in open-source implementations of mobile cores, born of the “blind trust” between network functions.
-
Servo 0.4: A Record Month and Ever More Correctly Rendered Sites
The Rust-based browser engine's update landed at the end of July with a record 558 commits and noticeably better compatibility with real-world websites.
-
Baidu Releases vLLM-Kunlun — a Plugin for Running LLMs on Kunlun XPU Accelerators
Baidu has open-sourced vLLM-Kunlun, a pluggable module that lets you run the popular vLLM inference engine on its own Kunlun XPU accelerators without patching vLLM's source code.
-
Amazon Attributes the Hijack of the npm Packages debug and chalk to North Korea's Sapphire Sleet
Ten months on, Amazon has tied the September 2025 attack on popular npm packages to the DPRK. The published evidence, however, turned out to be thinner than the bold claim.
-
GCC Will Reject Any Legally Significant AI-Generated Contribution — Except Tests
The GCC steering committee has approved a policy barring “legally significant” code generated by language models from the compiler. The only exception is for test cases.
-
ARCTIC Ships a Fan Controller with Ready-Made Linux Kernel Support
ARCTIC has not only released an inexpensive 10-channel fan controller, but also prepared an open-source driver for it in advance—one that has already landed in the mainline Linux kernel.
-
A $500 Fine-Tuned 9B Model Beat Top AIs at Catalog Review
The authors of Fermisense claim that an open 9-billion-parameter model, fine-tuned with reinforcement learning, outperformed leading proprietary models on a specific task and turned out to be dozens of times cheaper to run.
- More