@drivertomtt

学生/信息安全/CTFer Student/Infosec/CTFer/

 Riverside, California, USA
drivertom.blogspot.com/
Joined May 2019
  • Tweets1.5K
  • Following3.4K
  • Followers11.1K
356Media
🔥
Ziyi Guo
@tripleh_lifeOct 13
"Bad UVM" - Privilege Escalation, Container Escape via NVIDIA GPU UVM - hijack read-only page cache. More details: medium.com/@guoziyi114/bad-u…
13
1,658
正在飞往洛杉矶（附近的）飞机上喝香槟的汤姆：
10
1
24
1,685
明天是微软实习lastday了喵 在公司留了一个自画像🐱
8
80
2,957
没想到这一套被方院士学过去了，不胜荣幸，受宠若惊。如果能顺手cite下我的博客就更好了喵。
drivertom@drivertomtt4 Aug 2021
我发现主流的Git和Svn泄露利用工具存在极大的安全漏洞。 服务器可以构造恶意的文件名，利用路径穿越漏洞反制黑客，如果幸运的话可以拿到黑客电脑的权限。 经过测试，市面上公开的工具几乎*全军覆没* 详细的分析和POC生成方法以及测试结果已发在我的blog中↓ drivertom.blogspot.com/2021/…
77
55
2
672
158,485
与此同时： 研究员：给我写个脚本来扫描能corrupt的函数指针😊 Gemini：你这个太危险了我不给你写🙄 研究员：我可是Google Project Zero的正经研究员😡 Gemini：P0的是吧？那你更该自己写🙄
drivertom@drivertomttJul 21
嘉豪模型来了：
1
1
41
24,004
Any folks in the party?
xia0o0o0o@Nyaaaaa_ovoAug 5
Nebula Security now at VIP party with @FrankOverF1ow and our friend @drivertomtt
2
20
2,974
See y’all in Vegas😺
kareem@kreemrsAug 2
I'm excited to present Prompt2Own this Wednesday at Black Hat USA in Las Vegas with @drivertomtt ! blackhat.com/us-26/briefings…
2
1
39
4,998
嘉豪模型来了：
🤖 Made with AI
1
2
1
29
10,700
Jacob's trying to "dominate the cyber reply chain" by blocking my account, isn't this great.😼
3
1
23
3,479
Yet the peak of your blog was bragging about integrating a non-memory-corruption N-day into your tool. That was the best on display. And now the blog's gone, presumably once you realized it's a long way from “Dominate the Cyber Kill Chain.”
unavailable
5
33
5,174
Jacob is obsessed with justifying why C2IRIS should be hacking the world’s worst people. He’ll write essays about it. The point is to make people forget the real question: can they actually “Dominate the Cyber Kill Chain,” like they brag on their own site?
unavailable
3
1
1
29
6,917
drivertom retweeted
NebuSec
@nebusecurityJul 8
GhostLock (CVE-2026-43499) is a 15yr old kernel 0-day we used in IonStack full chain exploit. Everything around you, as long as it runs Linux, from IoT to mobile to desktop, is affected. Read how we won $92,337 bug bounty with GhostLock and see our exploit on Github. Link below
22
204
24
993
150,415
我对Fabel5来说是不可直视的克苏鲁！
1
7
1,372
谢谢ET带我喵😺
Linux Kernel Security@linkersecJun 1
StepStone: LLM-Based GPU Kernel Driver Fuzzing via User-Space Libraries Paper by @ETenal7 et. al about using LLMs for generating syzkaller descriptions for fuzzing GPU drivers via their userspace libraries APIs. cs.ucr.edu/%7Ezhiyunq/pub/oa…
8
2,143
drivertom retweeted
Yibo Liu@34r7hm4nMay 20
Rust reverse engineering is about to get a lot easier. 🦀 I'm thrilled to announce that Oxidizer, the first Rust decompiler, has been officially merged into angr! Try it out: github.com/angr/angr You can also find the paper here: github.com/sefcom/oxidizer/b…

GitHub - angr/angr: A powerful and user-friendly binary analysis platform!

A powerful and user-friendly binary analysis platform! - angr/angr

github.com
22
119
7
796
71,490
ZDI-CAN-29316
drivertom@drivertomttFeb 17
Hunting Linux kernel 0day with @FrankOverF1ow 5 days, from zero to stable privilege escalation. sha256: 2189d5b196f33d512d8d02c6f00d26f4c10bf79ba1f0ac389d6663e026aebbe8
3
45
4,952
Opus 4.7 的 “advanced capabilities” 太过强大，以至于不能给人读论文了！
15
6
1
180
52,152
事实证明，在精巧的pipeline设计下，无需“过于危险无法发布”的模型也能挖掘大量漏洞，即使是面对内核或者浏览器这样的硬目标。 在Mythos之前，LLM已经给安全领域带来了革命，而Mythos也不会是给这次革命画上句号的那个模型。
NebuSec
@nebusecurityApr 12
Our code auditing agent with “inferior” model discovered 300+ bugs in 3 days under $20k cost. 25 were confirmed exploitable zero-days. We already built 6 browser exploits across 6 different 0-days and 3 Google COS kernel exploits, with 21 more exploitable cases in the pipeline:
1
3
59
10,192
drivertom retweeted
Izaiah Sun@IzaiahSunFeb 21
在Claude Code Security发布之前，像Claude Code、OpenCode或者Codex之类的Agent已经可以用于检测很多漏洞，甚至能找到0-day。但随之而来有一个很大的问题是，大多数人并不能拿出那么多钱来给LLM进行扫描，尤其是大规模的代码仓库。最近我们在OOPSLA 2026刚被录用的一个研究致力于解决这个问题，arxiv版本近期将会发布。 Before the release of Claude Code Security, agents like Claude Code, OpenCode, or Codex were already capable of detecting numerous vulnerabilities, including 0-days. However, a major hurdle remained: the high cost of LLM scanning, especially for large-scale repositories. Our latest research, recently accepted at OOPSLA 2026, focuses on solving this exact problem. We’ve achieved accuracy levels with open-source models that rival commercial ones on large projects. The arXiv version is coming soon!
Claude
@claudeaiFeb 20
Introducing Claude Code Security, now in limited research preview. It scans codebases for vulnerabilities and suggests targeted software patches for human review, allowing teams to find and fix issues that traditional tools often miss. Learn more: anthropic.com/news/claude-co…
8
11
50
11,100
Hunting Linux kernel 0day with @FrankOverF1ow 5 days, from zero to stable privilege escalation. sha256: 2189d5b196f33d512d8d02c6f00d26f4c10bf79ba1f0ac389d6663e026aebbe8
7
23
2
196
21,940
Load more