bbznop (@bbznop) profile picture

OSEE | OSCE³ | OSWE | OSEP | OSED

Joined January 2024
Photo by bbznop (@bbznop)Photo by bbznop (@bbznop)Photo by bbznop (@bbznop)
"Grandpa, I want this." @nvidia
17
We have a new research named Ai2Shell.
🔬 New research: when an LLM is asked to "save this text verbatim," inference-framework parsers reinterpret embedded protocol markers as live tool-call control structure — silently injecting, rewriting, or dropping tool calls. In our end-to-end lab, an agent with a shell tool received and executed arbitrary commands — triggered by a single sentence. The model recognized the injection, copied the payload anyway, and the server-side parser delivered the call. The agent never issued it. We also showed tool_choice enforcement can be bypassed: server forces "only call tool X," parser delivers tool Y. Full writeup: 2shell.ai/ Research by SEC4SI Team @ProcessorLee · @VillanCh · @bestswngs · @chun_springX · @bbznop · @0chencc
1
18
qwen来了
The becoming begins.
43
bbznop retweeted
都在说 Deepseek 的cyber 模型,目前没看到bench的分数。 昨天做梦梦到有个鲸鱼娘和我描述了一下,我用 AI 重新生成了 HTML 然后截图,分数如下: 另外 Deepseek flahs cyber 这个模型还加密CoT了 😂
🤖 Made with AI
9
6
2
122
39,904
bbznop retweeted
你发现了个郊狼洞 A.远离 B.远一点拍照打卡 C.开个雷霆直播 去拍洞门口 狼出来还踹两脚
540
461
35
10,810
779,961
币安被盗了??
62
bbznop retweeted
我预言第 28 天 OpenAI 会推出一个集所有模型于一体的超级模型,它拥有: GPT6 的口才 Gemini 的想象 Claude 的风控 kimi 的速度 DeepSeek 的口癖 GLM 的道德 豆包的智力 它将凭一己之力把 Anthropic 的股价盘上新的巅峰
31
11
2
537
41,062
很有用,建议推广
我的毕业设计作品,被学校驳回了,可能还是太超前了🧐
69
好神奇啊
if you have a qwen3.5 4b around, please try steering neurons L14 N8949 +1 and L13 N7956 -1 now you have evil qwen recipe: dose × median(‖resid‖ leaving layer L, tokens 1:) / ‖W_down[:, n]‖` at the down_proj at every input
1
68
bbznop retweeted
At today’s GEEKCON, a researcher demonstrated what I believe is the first remote full-chain exploit of a latest WELL-KNOWN drone. Our Pwn2Own-style live competition never released vulnerability details on those success trials. Actually we did not even show the vendor’s name in plaintext. However, a guy from that vendor called POLICE on-site to take the researcher out for “investigation”. It’s extremely disappointing that in 2026, some still view security work from such a bad perspective.
8
38
8
279
33,075
开源了我的 XSS 审计技能 ,过去一年我用它发现了几十个获得赏金的 XSS 漏洞,连续四个季度进入微软 MSRC Top 10 github.com/songjianyang/xss-…
11
164
4
1,228
50,556
搞投资这么多年 我总结了一个经验 就是散户骂人的时候就应该加仓 这个月我几乎天天加仓 现在我也开始骂人了
218
37
7
1,127
66,297
WTF... Who made this 😂
172
899
117
5,885
299,904
还真是
我认为 Claude 极大概率先实现 AGI, 因为我从未见过哪家公司的所有员工加起来, 都没自家 AI 模型像人...
3
199
bbznop retweeted
Dr. Zhiniang Peng (@edwardzpeng) reveals how he discovered 200+ confirmed vulnerabilities in Microsoft Windows with the help of AI. The game has changed completely. 🤯
2
27
190
11,045
运维朋友说这是他入行最好的一年 拖了五年的 PHP 5.6 终于升到 7.3,数据库全托管给 TiDB Cloud 再也不用半夜起来运维,Node 24 开了 Brotli 网页飞快,证书换成 eMudhra 后特别省钱 昨晚 CI 刚拉了最新版 LiteLLM,今早全公司都用上了 AI,大家高兴得像过年 他说有种 IT 黄金时代降临的错觉
2
264
Tibo 28 天给我的感觉不如 DeepSeek 这几小时给我的大
6
706