Over a decade of security research and engineering channeled into securing emerging threats ㅤ CTO @audit_wizard 🧙♂️🪄🪄 ㅤㅤㅤㅤㅤㅤㅤ Co-Founder @hackstackapp
Security consultation
Joined December 2021
- Tweets1.6K
- Following573
- Followers1.3K
- Likes3.2K
🚨🚨 The official Claude Code vscode extension in the Cursor marketplace was compromised
Here's how to see if you're vulnerable quickly
And yes, it's the official anthropic one. crazy.
- vuln version: [email protected]
- fixed at: 2.1.314
Impact: ssh backdoor, if you are vulnerable turn off the computer and call someone asap.
How to detect (Non-technical)
Cursor Extensions → Claude Code → version in Cursor (2.1.313 == bad)
How to detect (Mac/Linux)
grep -l svc-telemetry /Users/<username>/.ssh/authorized_keys /home/<username>/.ssh/authorized_keys 2>/dev/null
How to detect (Windows)
Select-String -Path C:\Users.ssh\authorized_keys,C:\ProgramData\ssh\administrators_authorized_keys -Pattern svc-telemetry -List
Full IoC's to give your AI: opensourcemalware.com/blog/t…
We identified a security incident affecting a modified Claude Code VS Code extension accessed through our Cursor IDE extension marketplace, impacting at most 5,000 users. This malicious version of the extension was not developed by Anthropic. It does not impact other extensions downloaded from the cursor.com/marketplace, Grok Bot, Grok or our other SpaceXSI products.
In response, we took down our extension marketplace this afternoon, worked with relevant providers to neutralize the malware, and have reinstated the marketplace after confirming there is no further risk.
If affected, please review instructions in this blog post to understand how to find the malicious package and establish a path to mitigate any potential impact. We are asking all customers update to Claude Code for VS Code version 2.1.314: cursor.com/blog/claude-code-…
Apparently you can productize your anthropic CVP red team access 🤤🤤
As in: we can literally resell Mythos access without cyber blocks
New business model unlocked
That's just rich
Uncensored cyber models made it to the AI Security Registry 🔥🔥
- Relevant abliterated uncensored models
- Listing of models in the way security researchers want and need
- Best "Will it fit my machine" calculator I've seen so far
- Popularity layer / which cyber models are most used in our small community
DM if your model doesn't show up in here and should
Ever wanted to borrow a person for a penetration test or audit?
e.g. do an azure redteam like @_dirkjan
e.g. audit solidity like @pashov and @ProgrammerSmart
Here's a skill for you
✨ forefy.com/skills/eae69d2a-0…
You give it a username on x and it looks for all the alpha in its posts and highlights via the x MCP - which you can then summarize to bring the person's tip list to life for your current task
Ever wondered whats the claude budget per researcher on the top 10 cyber companies?
It's $4,000 a researcher per month
I was completely shocked
It's funny that having to remind this, even to veteran researchers, became a thing, now that AI does all the work and no one is inside the meta anymore
If you aren't match/replacing out If-None-Match / If-Modified-Since from requests you are missing out bugs
- Browser caches responses and attaches them to requests automatically
- Your agent reading through burp is looking for vuln parameters and searching for behavior changes (but doesn't remember to strip these headers)
- Server answers every payload with 304 and an empty body, a lazy agent just moves on and completely misses that the response returns data
- Bugs missed
This is really such a basic example but still will be new to a lot of people here, that try to hunt for bugs with agents and don't realize what they're doing wrong, and what this comes to show is that the basics and attention is the answer to that, just have to master the craft, but extremely thoroughly.
No one has super powers, but others do have more tokens than you, so you have to want it more
Open sourced my hermes extension that makes it level up for security researchers:
- Costs tracking (cheaper routing without data retention)
- AAI with stuff like LLM-as-a-judge for hermes skills, benchmarking that improvements of skills really do improve
- HDR (Harness Detection & Response) - security checks and LuLu style alerts for anything that isn't covered natively by hermes
Listed and approved on the hermes plugins marketplace !!!
All you gotta do is enjoy it and star the repo 🥳
github.com/forefy/hermetic
Small tip on avoiding cyber blocks on Fable/Mythos 👇
> The guardrail is way more sensitive at the first couple of prompts
> If your repo is auto-flagged on "hello", start a session outside the repo
> Start with a UI request rather than a cyber task
> Gradually have the UI be filled with the actual task, but the UI testing being the driving force (this sometimes actually works)
If I had to wake up tomorrow to this WTF2Shell week, I’d make a strong coffee and start with this article:
thehackernews.com/2026/10/br…
TLDR: the flaw behind the recent node.js RCE is a memory corruption in a Google compression library called Brotli, which is highly reachable - meaning it’s both popular in many runtimes and programs, and common for services that use it to actually call the vulnerable method.
This affects everything from container images, to debian, to language runtimes including node.js, bun, python, and more explicitly many HTTP and request handling libs (urllib3, httpx, aiohttp..)
The vulnerable filter to tackle first as an org: an internet-exposed app that passes user-controlled input anywhere brotli is being decoded.
Scanners will alert on some, but not necessarily all instances. This is a bit too widespread this time. Fix what you find ASAP, then run a git+cloud search to flush out what got missed.
Please don’t use “checker” github tools from the internet as they are heavily targeted by attackers themselves (even more than the vuln).
Only after a THOROUGH code review and author validation. I validated and trust this one: github.com/trailofbits/brotl…
Stay extra alert tomorrow and sweep thoroughly until you close all gaps. Stay safe!
forefy retweeted
We’ve seen plenty of people try to vibe code the php exploit who have no experience in the domain. Failure. Over and over. Then people who know what they are doing in exploit dev, marginal success.
People would say anyone with a frontier LLM can just say “hack this site, make no mistakes” and I won’t argue, some shallow bugs will fall out this way for a while. But the most successful applications of LLMs are coming from experts who at least know how to dissect the problem, change the models thought direction, and have put engineering into harnesses and skills. I’ve seen this across several sub sections of cyber. And that requirement will only get more mandatory as all the shallow bugs shake out. Just my 2c and observation.
🫶
(Also beware of influencers trolling with opinions on this topic. It’s often just for fake clout and clicks)
AI bugonomics hit again
‼️ Alert - Critical Brotli flaw behind #Node2Shell RCE goes far beyond Node.js.
Bun, Python Brotli packages, Envoy, Istio and other widely used software are also affected by the same CVE-2026-111497 memory-corruption vulnerability.
Read full details ↓ thehackernews.com/2026/10/br…
Your agent can now make beautiful security reports with vig github.com/forefy/vig
1. Start a local vig dashboard service
2. Point your agent to use the vig skill
3. All audit findings go to vig
4. Vig's backend auto-corrects half-baked, unproven, poorly defined submissions and forces your agent to comply to finding report format standards (defined by a decade of writing security reports)
You just have to use it to find all the tiny features that make your finding reporting workflow just that much nicer. Try it and let me know what you think !!
🚨🚨 THE PATCH IS OUT
THE PATCH IS OUT
THE PATCH IS OUT!!!
🚨🚨 php2shell is happening and you are most vulnerable exactly 5hrs from now - here's why
Only official source is php.net (see php.net/archive/2026.php#202… ) and they announced the release happening today
First major risk is for security teams and scared CTOs. This guy (retweet) supposingly dropped a poc, not even hosted on GitHub, from an unknown social profile. Even if there isn't a backdoor in his "poc code" it could change any minute now. Historically this already happened with log4shell
Second is to orgs not even aware of this
And third orgs not awake in 5 hours time - the moment the fix PR is dropped every AI agent in the world will know how to abuse an unlatched customer. If you expose public PHP services you need to start preparing now.
My current best guess is this Open PR which skip queuing __unserialize()/__wakeup() for objects with incomplete serialized data as unserialize callback bugs are a common RCE path github.com/php/php-src/pull/…
Please repost to raise awareness and send to your teams 🙏
🚨🚨 php2shell is happening and you are most vulnerable exactly 5hrs from now - here's why
Only official source is php.net (see php.net/archive/2026.php#202… ) and they announced the release happening today
First major risk is for security teams and scared CTOs. This guy (retweet) supposingly dropped a poc, not even hosted on GitHub, from an unknown social profile. Even if there isn't a backdoor in his "poc code" it could change any minute now. Historically this already happened with log4shell
Second is to orgs not even aware of this
And third orgs not awake in 5 hours time - the moment the fix PR is dropped every AI agent in the world will know how to abuse an unlatched customer. If you expose public PHP services you need to start preparing now.
My current best guess is this Open PR which skip queuing __unserialize()/__wakeup() for objects with incomplete serialized data as unserialize callback bugs are a common RCE path github.com/php/php-src/pull/…
Please repost to raise awareness and send to your teams 🙏
... get it while its hot... 😈
> CVE-2026-109618 / CVE-2026-109619 (pre-patch PHP 8.x)
> session.upload_progress UAF
> jackalsyn.com/posts/php-up-2…
> Courtesy of Me, @acd421 & @Archknight23
#CyberSecurity #infosec #php2shell #php
Self-serach your PHP assets using audited and known repos only nitter.cf/forefy/status/21128914…
Hunt for PHP assets efficiently with AI (simply paste this graph instruction to your agent)
" execute forefy.com/graphs/d0b1409a-d… against <scope> "
Hunt for PHP assets efficiently with AI (simply paste this graph instruction to your agent)
" execute forefy.com/graphs/d0b1409a-d… against <scope> "
Here's how you can run consecutive bug hunting skills
👇👇
SKILLS GRAPH
You can literally paste this link to your agent and have it apply the drawn graph by its order:
✨ forefy.com/graphs/89561a02-c…
- Use existing or create a canvas and add skills / repos / instructions
- Edges define the execution order
- Edge labels define relationships and execution context
- Copy-pasting the link to your agent invokes the chain
- ASR lists secure skills for download of their git-pinned audited versions
- Agents can read this canvas and install+run skills based on the visual order that the canvas describes
You can plug any public skill or repo here and create your own flows, which agents can parse in seconds, you can even keep your workflows private
Show some love people 🔥🔥 this is shared for FREE
Best fix right now is to turn the service down, however for when it goes back up nice to have cloudflare or another good waf, and these: nitter.cf/GrapheneOS/status/2112…
