Pinned Tweet
OpenAI has announced a partnership with Hacktron to investigate vulnerabilities in its systems. Our researchers will work with @OpenAI's engineers to prioritize testing of connections with third-party services, with broader frontier-model access through Daybreak Red.
We're excited to deepen our partnership with OpenAI, combining our offensive security expertise with their frontier models to help secure OpenAI and the broader software ecosystem.
openai.com/index/partnering-…
nginx2root: five nginx bugs, each leading to code execution only under a specific configuration. CVE-2026-123007 affects unpatched nginx with HTTP/2 enabled and has been in nginx since 2015.
Fixed on 18 November in 1.31.9 and 1.30.7.
Who’s actually exposed: hacktron.ai/blog/nginx2root?…
To check CVE-2026-123007, run nginx -v, then confirm the binary has HTTP/2 (nginx -V) and your config turns it on (nginx -T).
Distributions backport fixes without changing the version, so compare the full package version with your distro’s advisory.
1
185
Java4Shell (CVE-2026-114938) is a heap overflow in OpenJDK’s accelerated Base64 decoder that can allow unauthenticated remote code execution. Fixes shipped on 17 November.
Who’s exposed, how to check, and what to update:
hacktron.ai/blog/java4shell?…
Calif’s demonstrated chains use corrupted array lengths to turn the overflow into a heap leak (JavaBleed), then build toward code execution.
The details depend on the application and heap layout. Remote grooming is sensitive to network latency.
1
196
Hacktron AI retweeted
You all need to patch this vulnerability. During an unrelated research effort we identified this vulnerability mid oct and reported to vendor and left to vendor to deal with ruby on patching this bug 🙃
About a month ago, I looked into Ruby and independently found bugs that make a universal pre-auth RCE on Ruby on Rails. In the current zeitgeist, people would refer to this as ruby2shell or rails2shell. However, @_sy1vi3 doesn't approve that.
Blog post: durendal.sh/blog/ruby-case-m…
Security teams have been asking, "Could the Binance breach happen to us?"
Binance had controls, policies, and passed audits. Yet none of that stopped an attacker from getting from one employee's machine to the funds.
Hacktron Red acts like a persistent adversary to find those paths before attackers do: hacktron.ai/red
Have a burning security question? Just @ Hacktron AI, now in Slack.
Check whether a new CVE affects your repos, ask how your code handles something, or save a decision straight to your threat model, all from the thread. It also triages findings, opens fix PRs, reviews pull requests and sets up pentests.
Get answers from your own code and threat model within minutes.
Try it yourself: hacktron.ai/docs/platform/sl…
hacktron website just got a fresh upgrade!
all thanks to the one and only @enscribe (enscribe.dev/gallery/). insanely talented goat. never seen an artist cook up things from scratch and turn them into absolute madness (he designed our old site too)
hacktron.ai/
Hacktron AI retweeted
Hacktron has reported high-impact vulns to us through our bug bounty program. Now we’re teaming up with @zeyu1337 and the Hacktron crew to hunt for more. openai.com/index/partnering-…








