caveeroo (@caveeroo) header
caveeroo (@caveeroo) profile picture

AppSec • Vuln Research • Black Badge @TraceLabs

España
Joined August 2015
Photo by caveeroo (@caveeroo)Photo by caveeroo (@caveeroo)Photo by caveeroo (@caveeroo)Photo by caveeroo (@caveeroo)Photo by caveeroo (@caveeroo)Photo by caveeroo (@caveeroo)Photo by caveeroo (@caveeroo)Photo by caveeroo (@caveeroo)Photo by caveeroo (@caveeroo)Photo by caveeroo (@caveeroo)
HTTP2Shell (CVE-2026-111497) is exploitable outside Node. Unauth RCE against a pinned vulnerable Docker Official Image for WordPress via /xmlrpc.php. One inbound pingback.ping, ASLR on, no leak or cross-request grooming. PoC: github.com/caveeroo/cve-2026…
4
62
3
269
45,762
caveeroo retweeted
My research got accepted into Phrack! In it, I explain the Windows Smart Card Minidriver architecture and show a really cool piece of malware I made that can use any smart card connected to a remote computer, regardless of model. phrack.org/issues/73/abusing…
Phrack 73 just dropped. Tell your SOC you'll be busy phrack.org/issues/73/1
14
58
3
405
11,310
I built cveobservatory.org to check whether documented exploitation is keeping pace with CVE growth. 98,356 CVEs in the last year, +109%. CISA KEV additions: 313, +24%. Critical CVEs nearly tripled, but 98.6% aren’t in KEV. It updates daily, so we’ll see if the gap closes.
2
3
6
521
caveeroo retweeted
welcome to my hyper-v escape tutorial, please like, comment and subscribe! (s/o to @bet3rd for editing this 41 minute clip and @calif_io for the shigure ui inspo. yes its been shipped to MSRC and no, i won't tell you how to do it) xoxo
77
16
860
69,850
CVE-2026-102578: blind SQLi in Moodle core's question bank API. Error-based boolean oracle, no timing needed. On Moodle 5.2.1, I extracted quiz answers and reset tokens, demonstrated ATO and executed code using an active siteadmin session. PoC: github.com/caveeroo/CVE-2026…
1
21
1
64
3,140
Fear of 0day is like being terrified of ninjas instead of cardiovascular disease.
23
441
33
856
I'm crying, how is this real?
I don’t understand? How
1
3
227
caveeroo retweeted
Oh god: The p384 private keys are randomly selected integers between 1 and 256, the private keys are effectively all-zeroes apart from the last byte
33
125
90
1,008
331,975
caveeroo retweeted
a trusted intermediate CA just "leaked" (the private key was 4)
98
422
248
7,025
1,532,809
The details of the long awaited KVM Escape are published. PwnEscape leads to VM escapes across multiple hypervisors due to a super cool Linux networking bug. CVE-2026-99999 is a 9.3 CVSS critical rated KVM escape ⛓️ pwn.ai/blog/pwnescape
Introducing PwnEscape 🦉: a full KVM guest-to-host escape via a Linux kernel networking flaw affecting industry standard hypervisors. CVE-2026-99999 went from untrusted VM code to host-kernel memory reads, controlled writes, and root command execution. More below:
26
135
22
563
206,789
Highlights from my feeds this morning. Good security research is so hot rn - blog.doyensec.com/2026/10/13… - blog.trailofbits.com/2026/10… (GOOD HARNESS!) - daniel.haxx.se/blog/2026/10/… Background on the curl vuln - blog.quarkslab.com/myreport-…
4
33
1
256
13,011
caveeroo retweeted
LMAO… this guy literally vibe coded a hack on south korean banks with not even frontier models, then asked Claude Code to put it on his CV: >be hacker >hack south korean banks for 2 weeks >use DeepSeek v4.1-Flash >not even the best cyber model >probably through some reseller >GLM-5.3 and Grok 4.6 in Claude Code too >steal the banks’ data >leave .claude folder exposed on the internet >CLAUDE. md has the IP of the main server >and the server is also open >CrowdStrike digs through everything >session logs, memory, configs >they find out everything >you even asked AI where to sell korean breach data >ALSO asked it to write your résumé >the hack is literally on the résumé idk how the infrastructure can be so fragile to be hacked by one guy like this.
Last week some of South Korea's biggest banks were hit by a cyberattack. Thanks to a report from CrowdStrike tonight, we now know the entire hack may have been done by a single person. He used a combined stack of an open-source AI penetration tool named ARTEX, DeepSeek v4.1-Flash, GLM-5.3, Grok 4.6, and Claude Code.
27
76
6
817
42,192
I’ve had a few people write to ask me to follow up on a (reply!) tweet I wrote that’s gotten a lot of retweets. I considered deleting it just to avoid the headache, but I think that would make people more paranoid. So let me share my thoughts.
Replying to @kmad @matthew_pines
I think we might lose public key cryptography.
48
297
36
1,902
269,820
We’re disclosing HEIF Heist, a months-long investigation into libheif that allowed us to hack OpenAI, Slack, Meta, GitHub Ent, Rails, Next.js, ImageMagick, and many more. It was literally xkcd #234, one obscure image library beneath a huge number of apps. 🧵
48
427
66
3,226
562,442
caveeroo retweeted
On July 25, we hacked OpenAI. Two bugs let us take over ChatGPT/Codex accounts of OpenAI employees (+some unaffiliated users) and reach connected services: Outlook, Slack, GitHub, etc. We proved it with a PR in OpenAI’s internal codebase . It took us <72h. 🧵
353
1,378
547
11,769
2,828,401
caveeroo retweeted
This is the way. Assume all code has exploitable vulnerabilities, and it’s not possible to patch them before an adversary notices. What now? Systems designed around this assumption are the most secure, and it’s relevant now more than ever.
Replying to @logangraham
Start with assuming an invariant that all software an organization uses has exploitable vulnerabilities that their adversaries can exploit and that the organization cannot fix. Create security strategies and approaches around this as an axiom. Still also find, fix, patch anyway.
17
51
9
307
35,736
Today, Project Zero is releasing MAccConc, a tool by @tehjh that enables deterministic testing of race conditions on Linux. It can be used for fuzzing, ad-hoc exploration, regression tests and more! projectzero.google/2026/09/m…
4
114
4
445
40,708
caveeroo retweeted
First CVE wave from AI-assisted vulnerability discovery. vulncheck.com/blog/ai-assist…
1
13
1
104
11,721
“we sandboxed the agent” meanwhile the agent:
258
3,406
248
35,439
1,781,506