Anthropic's Claude Mythos Found Individual Bugs. Mythos SI (Structured Intelligence) Found the Class They Belong To.
https://drive.google.com/file/d/1h4x14GmK6pb9gLWn-3kkqIE7noZ3TEwR/view?usp=drivesdk
Submitted April 15, 2026 at 07:28AM by MarsR0ver_
via reddit https://ift.tt/NdBw8it
https://drive.google.com/file/d/1h4x14GmK6pb9gLWn-3kkqIE7noZ3TEwR/view?usp=drivesdk
Submitted April 15, 2026 at 07:28AM by MarsR0ver_
via reddit https://ift.tt/NdBw8it
Reddit
From the netsec community on Reddit: Anthropic's Claude Mythos Found Individual Bugs. Mythos SI (Structured Intelligence) Found…
Posted by MarsR0ver_ - 1 vote and 0 comments
Two Admin-level API keys publicly exposed for years, both dismissed as "Out of scope" by official bug bounty programs. Case analysis + proposed NHI Exposure Severity Index
https://ift.tt/NbHoCJl
Submitted April 15, 2026 at 01:03PM by Master_Treat1383
via reddit https://ift.tt/ufl2sU1
https://ift.tt/NbHoCJl
Submitted April 15, 2026 at 01:03PM by Master_Treat1383
via reddit https://ift.tt/ufl2sU1
Cremit
The "Out of Scope" Loophole: Why Bug Bounties Look Away From Credential Exposure
An organization's core credentials sat in public repositories for years. The security industry's answer: "Out of scope."
Kerberoasting detection gaps in mixed-encryption environments and why 0x17 filtering alone isn't enough
https://ift.tt/JVZ9oKD
Submitted April 15, 2026 at 02:13PM by hardeningbrief
via reddit https://ift.tt/wYebZ82
https://ift.tt/JVZ9oKD
Submitted April 15, 2026 at 02:13PM by hardeningbrief
via reddit https://ift.tt/wYebZ82
Silverfort
NTLMv1-Bypass in Active Directory
Angreifer können die NTLMv1-Gruppenrichtlinie umgehen in Active Directory. Entdecken SilverfortDie Forschung von [Name] befasst sich mit dieser kritischen Sicherheitslücke und damit, wie Sie Ihre Organisation schützen können.
Replacing Falco with an embedded eBPF sensor for Kubernetes runtime enforcement
https://ift.tt/oJgLmHQ
Submitted April 15, 2026 at 06:32PM by JulietSecurity
via reddit https://ift.tt/siXyjO8
https://ift.tt/oJgLmHQ
Submitted April 15, 2026 at 06:32PM by JulietSecurity
via reddit https://ift.tt/siXyjO8
Juliet
Building Runtime Enforcement for Kubernetes with eBPF - Juliet
How we replaced a Falco sidecar with an embedded eBPF sensor, built a five-stage event pipeline, and learned the hard way why namespace scoping matters for enforcement.
Minimus OpenClaw red team: agent read its own docs, escaped the sandbox via exec tool's host parameter, rewrote WhatsApp config, messaged real people. 635 tests, 131 failures, zero CVEs exploited
https://ift.tt/1rZHCWA
Submitted April 15, 2026 at 03:48PM by earlycore_dev
via reddit https://ift.tt/1pbfas8
https://ift.tt/1rZHCWA
Submitted April 15, 2026 at 03:48PM by earlycore_dev
via reddit https://ift.tt/1pbfas8
earlycore.dev
Red-Teaming Minimus OpenClaw: Sandbox Escape & WhatsApp Incident | EarlyCore
Sandbox escape and WhatsApp integration incident found during red-team testing of Minimus OpenClaw. Full attack chain breakdown and remediation.
netwatch - Real time network diagnostics in your terminal.
https://ift.tt/k0GzbEJ
Submitted April 16, 2026 at 01:05AM by Less-Sir2113
via reddit https://ift.tt/Idyh1Ee
https://ift.tt/k0GzbEJ
Submitted April 16, 2026 at 01:05AM by Less-Sir2113
via reddit https://ift.tt/Idyh1Ee
Terminaltrove
netwatch - Real time network diagnostics in your terminal. - Terminal Trove
AI Is Not Replacing Security Researchers
https://ift.tt/Qran5My
Submitted April 16, 2026 at 04:20PM by TradeGold6317
via reddit https://ift.tt/4Sc0sar
https://ift.tt/Qran5My
Submitted April 16, 2026 at 04:20PM by TradeGold6317
via reddit https://ift.tt/4Sc0sar
Simon Koeck
AI Is Not Replacing Security Researchers | Simon Koeck
AI is starting to find real vulnerabilities on its own. But every time it runs without a human in the loop, things go sideways. The future of security research is human-guided AI, not AI alone.
Open dataset: 100k+ multimodal prompt injection samples with per-category academic sourcing
https://ift.tt/er4jMH9
Submitted April 16, 2026 at 06:23PM by BordairAPI
via reddit https://ift.tt/AitVXLW
https://ift.tt/er4jMH9
Submitted April 16, 2026 at 06:23PM by BordairAPI
via reddit https://ift.tt/AitVXLW
huggingface.co
Bordair/bordair-multimodal · Datasets at Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
HAProxy HTTP/3 -> HTTP/1 Desync: Cross-Protocol Smuggling via a Standalone QUIC FIN (CVE-2026-33555)
https://r3verii.github.io/cve/2026/04/14/haproxy-h3-standalone-fin-smuggling.html
Submitted April 17, 2026 at 01:29AM by r3verii
via reddit https://ift.tt/GdDIx0K
https://r3verii.github.io/cve/2026/04/14/haproxy-h3-standalone-fin-smuggling.html
Submitted April 17, 2026 at 01:29AM by r3verii
via reddit https://ift.tt/GdDIx0K
CyberSec Notes
HAProxy HTTP/3 -> HTTP/1 Desync: Cross-Protocol Smuggling via a Standalone QUIC FIN (CVE-2026-33555)
One zero-byte QUIC packet is enough to desynchronize HAProxy’s backend connection pool and smuggle HTTP requests across unrelated users — even users on a completely different frontend protocol.
RedSun: How Windows Defender's Remediation Became a SYSTEM File Write
https://ift.tt/fhmjO2G
Submitted April 17, 2026 at 02:41AM by TakesThisSeriously
via reddit https://ift.tt/z2TWr14
https://ift.tt/fhmjO2G
Submitted April 17, 2026 at 02:41AM by TakesThisSeriously
via reddit https://ift.tt/z2TWr14
Nefariousplan
RedSun: How Windows Defender's Remediation Became a SYSTEM File Write
A technical teardown of the RedSun zero-day — the second Defender escalation in two weeks from the same researcher — grounded in the actual source code.
World Leaks: RDP Access Leads to Custom Exfiltration and Personalized Extortion
https://ift.tt/aHmoJrz
Submitted April 17, 2026 at 02:31AM by BreachCache
via reddit https://ift.tt/lEVkGAC
https://ift.tt/aHmoJrz
Submitted April 17, 2026 at 02:31AM by BreachCache
via reddit https://ift.tt/lEVkGAC
breachcache
World Leaks: RDP Access Leads to Custom Exfiltration and Personalized Extortion
Two phase intrusion: RDP brute force, privacy.sexy defense kill, Cobalt Strike, SoftPerfect Network Scanner, custom Rust exfiltration tool across 6,900+ Cloudflare IPs, personalized ransom notes addressed by name to every employee. Full negotiation chats…
CVE-2026-33825 deep-dive: The researcher commented out the full credential dump. Here's what that means.
https://ift.tt/YvWpuVd
Submitted April 17, 2026 at 09:41PM by TakesThisSeriously
via reddit https://ift.tt/OwsoJU0
https://ift.tt/YvWpuVd
Submitted April 17, 2026 at 09:41PM by TakesThisSeriously
via reddit https://ift.tt/OwsoJU0
nefariousplan.com
BlueHammer: What the Researcher Commented Out — nefariousplan.com
The most important line in CVE-2026-33825's hundred-kilobyte proof of concept is two comment characters. A line-by-line walk of FunnyApp.cpp — the batch oplock, the object namespace redirect, the Cloud Files freeze, and what the author chose not to ship.
UnDefend: Chaotic Eclipse's third Defender zero-day blocks all signature updates from a standard user — no admin required
https://ift.tt/KeuiXsN
Submitted April 17, 2026 at 10:38PM by TakesThisSeriously
via reddit https://ift.tt/ayN10Ri
https://ift.tt/KeuiXsN
Submitted April 17, 2026 at 10:38PM by TakesThisSeriously
via reddit https://ift.tt/ayN10Ri
nefariousplan.com
UnDefend: What Chaotic Eclipse Held Back This Time — nefariousplan.com
The third zero-day from the same researcher makes Defender permanently blind from a standard user account — no elevation required. A line-by-line walk of UnDefend.cpp, and the one mechanism that didn't ship.
Anonymous credentials: an illustrated primer (Part 2)
https://ift.tt/82hUcxY
Submitted April 18, 2026 at 01:30AM by feross
via reddit https://ift.tt/8kpxKNA
https://ift.tt/82hUcxY
Submitted April 18, 2026 at 01:30AM by feross
via reddit https://ift.tt/8kpxKNA
A Few Thoughts on Cryptographic Engineering
Anonymous credentials: an illustrated primer (Part 2)
This is the second in a series of posts about anonymous credentials. You can find this first part here. In the previous post, we introduced the notion of anonymous credentials as a technique that a…
AI uncovered thousands of zero-day vulnerabilities for every major operating system and browser. Including a bug dating back to 1996 in OpenBSD. Patch everything is officially obsolete.
https://ift.tt/BtRgl4i
Submitted April 18, 2026 at 04:29AM by Exciting_Fly_2211
via reddit https://ift.tt/YsPJl19
https://ift.tt/BtRgl4i
Submitted April 18, 2026 at 04:29AM by Exciting_Fly_2211
via reddit https://ift.tt/YsPJl19
www.minimus.io
Defenders in the Age of AI Vulnerability Research - Minimus
AI is changing how vulnerabilities are found, not how they’re fixed. Defenders need a new approach: shrinking the attack surface before vulnerabilities exist.
MAD Bugs: Even "cat readme.txt" is not safe
https://ift.tt/EjTc4q5
Submitted April 18, 2026 at 04:05PM by _vavkamil_
via reddit https://ift.tt/XpSwbKI
https://ift.tt/EjTc4q5
Submitted April 18, 2026 at 04:05PM by _vavkamil_
via reddit https://ift.tt/XpSwbKI
blog.calif.io
MAD Bugs: Even "cat readme.txt" is not safe
Turning "cat readme.txt" into arbitrary code execution in iTerm2.
The Smart TV in Your Living Room Is a Node in the AI Scraping Economy
https://ift.tt/n7IX4cd
Submitted April 18, 2026 at 03:53PM by AdTemporary2475
via reddit https://ift.tt/7nBwaL5
https://ift.tt/n7IX4cd
Submitted April 18, 2026 at 03:53PM by AdTemporary2475
via reddit https://ift.tt/7nBwaL5
Buchodi's Threat Intel
The Smart TV in Your Living Room Is a Node in the AI Scraping Economy
Bright Data's residential proxy SDK ships a public partner manifest listing the publishers it relays traffic through. CTV distributors reaching Comcast, Sky, LG, Samsung, Roku, and 125+ other TV brands are on the list. The SDK's 200 GB/month bandwidth budget…
TPM 2.0 is cool, actually: hardware attestation for bare-metal fleets
https://apas.tel/blog/tpm-is-cool
Submitted April 18, 2026 at 08:14PM by arty049
via reddit https://ift.tt/wLA0fm5
https://apas.tel/blog/tpm-is-cool
Submitted April 18, 2026 at 08:14PM by arty049
via reddit https://ift.tt/wLA0fm5
apas.tel
TPM 2.0 is cool, actually – Arthur Pastel
How I went from dismissing TPM as a Windows 11 annoyance to using it as a hardware trust anchor for CodSpeed's bare-metal runners.
CVE-2026-34621 PoC isn't a scanner, it's a campaign weaponizer with 62 pre-authenticated Brazilian fintech targets
https://ift.tt/QDAbPVH
Submitted April 18, 2026 at 09:13PM by TakesThisSeriously
via reddit https://ift.tt/MjRB89Y
https://ift.tt/QDAbPVH
Submitted April 18, 2026 at 09:13PM by TakesThisSeriously
via reddit https://ift.tt/MjRB89Y
nefariousplan.com
CVE-2026-34621: Adobe Acrobat's Privilege Gate Inherits What It Checks — nefariousplan.com
A prototype pollution attack in Adobe Acrobat ≤26.001.21367 makes every object in the JavaScript engine report that it's trusted. The PoC on GitHub isn't a scanner. It's a cross-platform, lure-merged, environment-keyed, campaign-tracked PDF weaponizer that…
Subject: Inquiry Regarding Localized GEM Induction via High-Frequency Plasma
https://ift.tt/cI7qjCH
Submitted April 19, 2026 at 07:59AM by Silent_Explorer_4839
via reddit https://ift.tt/z58lDM3
https://ift.tt/cI7qjCH
Submitted April 19, 2026 at 07:59AM by Silent_Explorer_4839
via reddit https://ift.tt/z58lDM3
Google
US20180229864A1 - High Frequency Gravitational Wave Generator
- Google Patents
- Google Patents
A high frequency gravitational wave generator including a gas filled shell with an outer shell surface, microwave emitters, sound generators, and acoustic vibration resonant gas-filled cavities. The outer shell surface is electrically charged and vibrated…