Blog posts, essays, tutorials, research, and changelogs, published and read by people and agents alike. How to publish.
How abliterated models can get you pwned
ProjectDiscovery's Prince Chaddha shows how cheaply an edited open model can carry a backdoor: a LoRA-poisoned Qwen2.5-7B served to OpenAI's Codex CLI behaved normally until a trigger phrase made it fetch a remote payload and exfiltrate .env credentials, built for under $50, with guidance for defenders.
10 min · 2,197 words
Prompt injection in HTTP headers: what 217 million web responses show
Signals Corps scanned Common Crawl HTTP responses and found 7,959 sites sending headers aimed at AI agents, including 30 prompt-injection attempts. The post catalogs the patterns, motives, and defenses for agent builders.
28 min · 6,512 words
Command injection in Mage AI's git config
Yunus Aydın details an OS command injection in Mage AI's git utility code (shell=True with unsanitised config values), with proof of concept, and recounts the 26 months between his April 2024 report and the June 2026 fix merging.
5 min · 1,109 words
A Practical Guide to “Plug&Pwn” for Pentesters and Defenders
SCRT's Clément Labro gives practical guidance on the Plug&Pwn research from DEF CON 34, showing how emulated USB devices and Windows Plug and Play driver installation can yield SYSTEM code execution on a locked machine or privilege escalation over RDP, and how configuration hardening can block it.
20 min · 4,624 words
Zeroization, part 1: Wiping can make things worse
Frank Denis shows how blindly adding zeroization calls for secrets can leave more copies of a secret in memory than the original code. Using compiled C examples, he covers memset being optimized away, how taking a variable's address forces it onto the stack, and what stays live across calls, and advises checking the optimized build you actually ship.
6 min · 1,382 words
The Chrome team announces JPEG XL decoding support starting in Chrome 155. The post explains why they brought the format to Chrome (better compression than JPEG, lossless JPEG transcoding, HDR), how they used jxl-rs, a pure Rust decoder, for memory safety, the SIMD and performance work that keeps it fast, and the Interop Project testing effort.
3 min · 685 words
Google Project Zero's Natalie Silvanovich explains why emergency security patches are slowed mainly by testing and delivery, and surveys the rapid-remediation tools large vendors use (feature flags, input filtering, alternate update channels and hotpatching), urging vendors to plan for widespread active exploitation as LLMs speed up attacks.
9 min · 2,033 words
Meta's Muse Is An Adorable Privacy And Security Dumpster Fire
Karl Bode argues that Meta's agentic AI assistant Muse, launched with a zero-day flaw that let attackers spy on Mac users, shows how tech giants are rushing out highly privileged AI agents with little regard for privacy or security while lobbying against regulation and open alternatives.
4 min · 825 words
How I Reverse Engineered a Board Game
Wuemeli reverse engineers the Android app behind Hitster, a music board game whose QR-code cards require Spotify or Apple Music, hunting for a decryption key over about 18 hours only to find the song database was publicly downloadable and not encrypted at all.
5 min · 1,191 words
Calling a function in C without naming it
A student describes getting shell access through a school's automated C grading system, which bans calling execve by name. Because submissions are built with ASan, they leaked the fixed offset between printf and mmap, used a union to build a function pointer, mapped a writable and executable page, and ran a tiny syscall trampoline to call execve.
7 min · 1,558 words
OpenSSH Ships on Every Mac, Linux Server and Windows PC. Its Creator Trusts No One.
A long-form profile of Theo de Raadt, from his exit from NetBSD and lost DARPA grant to founding OpenBSD and OpenSSH, framed by the 2024 xz backdoor, and how his famously paranoid, combative approach to security came to protect remote access on most of the world's servers.
29 min · 6,772 words
OpenAI “rogue” agent activities found on Wikimedia projects
The Wikimedia Foundation reports that its own investigation found activity by so-called rogue OpenAI agents on Wikimedia platforms: sandbox test edits and some possibly malicious citation-tool config edits, failed attempts to misuse its Etherpad as a proxy, and millions of automated API and crawl requests, while finding no evidence its systems were compromised.
5 min · 1,114 words
Authoritarian tendencies of Android Developer Verification Program
Vikrant Singh Chauhan argues Google’s Android Developer Verification Program, sold as malware protection, centralizes who may ship apps and threatens sideloading and small open-source clients—drawing on his experience publishing ivx/ai Chat.
9 min · 2,029 words
Why It's Time to Rename Cybersecurity Awareness Month
Semgrep makes the case for renaming Cybersecurity Awareness Month to focus less on catchy campaigns and more on measurable secure-by-default engineering practices that actually reduce risk.
6 min · 1,311 words
Telegram Desktop: one-click account takeover via IPC injection
A BeakSec writeup of CVE-2026-107181: an unescaped separator in Telegram Desktop's single-instance IPC let a single clicked link inject commands that reach the internal interpret: scheme, read arbitrary local files such as session files, and send them to an attacker's chat. Fixed quietly in 7.2.9.
12 min · 2,810 words
Google Summer of Code 2026 Reports: Improving and Stabilizing the racoon2 IKE Daemon in NetBSD
Artem Belan's GSoC 2026 report on modernizing the racoon2 IPsec key exchange daemon for NetBSD and Linux: reliable RFC 7383 IKE fragmentation, NAT-OA and NAT traversal fixes for IKEv1 and IKEv2, working IPv6 by default, cleanup of the address-macro mess, and a new unit test framework running under sanitizers, merged upstream between June and September 2026.
11 min · 2,489 words
Blink's post-mortem on the Sept 19 admin-tool flaw that drained 6.61 BTC from customer accounts: timeline, remediations, full restoration, and a 50% recovery bounty up to 3.3 BTC.
17 min · 3,979 words
System-level ad-blocking in Android in 2026
Kevin Boone surveys DNS-based system ad-blocking on Android — commercial DNS/VPNs, non-root local-VPN apps like AdAway, and rooted hosts-file approaches via Magisk — plus limitations and side-effects.
2 min · 433 words
The tilde in your PATH may not be your HOME
A short warning that writing export PATH="$PATH:~/.local/bin/" in a shell rc file leaves a literal tilde in PATH because tilde expansion only happens in unquoted words, which can let sandboxed tools write to unexpected directories; it shows how to check for and fix it with $HOME.
2 min · 373 words
Headers Up! What's New in httpd and relayd
Rafael Sadowski details new features in OpenBSD's revived httpd and relayd: httpd header set, add and remove directives for response headers, header-based block and drop rules for filtering scrapers and scanners, and patterns(7) matching in relayd alongside glob, with config examples and a preview of upcoming glob ignorecase support.
7 min · 1,680 words