손지훈

Laugh Now, Try Later..

[BioMan🪙 @ganziboy11]

트럼프 “미국은 비트코인 전략적 비축을 할거다”

(근데 돈주고 살 계획은 없다)

압수해서 코인을 비축하고 있는 미국 근황입니다 ㄷㄷ

@realDonaldTrump

[GitHubDaily @GitHub_Daily]

做安全渗透测试时，工具一个个手动跑，扫描完了再分析漏洞，流程繁琐且容易遗漏，效率极其低。

偶然看到 Pentest Swarm AI，号称第一个真正基于「群体智能」架构的开源渗透测试工具，不是简单的多智能体排队干活。

各个智能体把发现写到共享面板上，其他智能体根据信息素权重自动决定下一步行动，攻击链是自然涌现的，不靠中央调度。

GitHub：https://t.co/HFgut0KWsz

内置了子域名发现、端口扫描、漏洞匹配、漏洞利用、报告生成等完整流程，一条命令就能启动整个蜂群对目标展开测试。

还提供了预设的实战剧本，覆盖漏洞赏金、外部攻击面管理、CTF 解题等场景，直接拿来用就行。

如果你在做渗透测试或参加 CTF 比赛，想让 AI 帮你自动化整个流程，可以关注下这个项目。

[보안프로젝트 @ngnicky]

버그 바운티 헌팅, 침투 테스트 및 공격적 보안 워크플로를 위한 엄선된 특수 AI 에이전트 프롬프트 모음 https://t.co/AVwnKYjaSl

[보안프로젝트 @ngnicky]

최근 ZeroPath Research를 통해 ProFTPD의 특정 모듈(mod_sql)에서 발생하는 심각한 보안 취약점(CVE-2026-42167)이 공개되었습니다.
이 취약점은 인증 우회는 물론, 특정 조건에서 서버의 제어권을 완전히 탈취할 수 있는 위험성을 내포하고 있습니다.
https://t.co/DHQA7HbiIP

[Amol Avasare @TheAmolAvasare]

For clarity, we're running a small test on ~2% of new prosumer signups. Existing Pro and Max subscribers aren't affected.

[심월 @mindmoon_108]

별 다른 공지도 없이 앤트로픽에서 이젠 20달러 플랜으론 클로드 코드 사용이 불가능 하게 되었습니다

클로드 코드를 사용하고 싶으시면 이젠 100달러를 지불하셔야합니다.

출처 : 텔레그램 AI MASTERS

[NullSecurityX @NullSecurityX]

Bug Bounty Full Course: Recon, Dorking, XSS/LFI, CORS & Open Redirect on Live Targets | YesWeHack

✅ Subdomain Recon
✅ Google Dorking
✅ XSS & LFI
✅ CORS Misconfiguration
✅ Open Redirect Mass Hunt

Perfect for beginners👇
https://t.co/1AUJ6O0W76

[1024 @1024DevHub]

Microsoft Edge 是下载 Google Chrome 浏览器排名第一的浏览器，😂

[64비트사령부⚡️ @64bitcoinkr]

현재 GPT 5.5 Pro 코딩 성능 근황

Codex 도 아닌 일반 GPT 5.5 Pro 대화 10번만으로 마인크래프트 비슷한 카피 게임을 만들어버림

만든 사람은 잘 하면 3번만으로도 될거같다고 함.

정말 충격적인건 코딩 전용인 Codex가 아니고 '일반 대화' 창에서 만든거라는 것임.

[Akhil Sharma 🧠 System Design @AkhilAiri]

@VivekIntel Noice. Do post about this tool too (it’s similar) - https://t.co/Av6mqNuU11

[Vivek | Cybersecurity @VivekIntel]

🚀 Pentest Copilot — AI That Actually Hacks (Legally)

Autonomous pentesting agent that connects to Kali, runs tools, analyzes output, and keeps attacking — no manual babysitting.

⚡ Highlights:
• Agentic loop (auto recon → exploit → iterate)
• 16 tools + 100+ capabilities
• Burp Suite + browser automation
• Parallel subagents
• Auto tool installation

🎯 Use cases: Pentest | Bug bounty | CTF

🔗 https://t.co/W4NpROzPYb

#CyberSecurity #Pentesting #AI

[The Hacker News @TheHackersNews]

Attackers are exploiting CVE-2024-3721 in TBK DVRs to deploy Mirai variant Nexcorium.

It spreads via old exploits and default creds, persists on devices, and launches DDoS attacks. EoL TP-Link routers are also being targeted via known flaws.

🔗 Read → https://t.co/gApGzKzd6V

[The Kobeissi Letter @KobeissiLetter]

HORMUZ UPDATE:

Iran has now turned back 20 vessels attempting to cross the Strait of Hormuz today and the US "blockade" has turned back a total of 23 vessels.

It appears we are now entering a complete shutdown of the Strait of Hormuz.

Prior to the US blockade, vessels from Iran and Iran's allies were permitted to sail through the Strait of Hormuz.

Now, under the US "blockade," the US is prohibiting vessels from accessing Iran's ports and coast.

We may be seeing the first ever complete shutdown of Hormuz.

More details to come shortly.

[The Hacker News @TheHackersNews]

⚠️ Researchers uncovered PowMix Botnet, active since Dec 2025.

Randomized C2 beaconing and phishing ZIP → LNK → PowerShell chains enable in-memory control and persistence.

RondoDox separately exploits 170+ flaws for DDoS and crypto mining.

🔗Read → https://t.co/JA2t5bNfBz

[金のニワトリ @gosrum]

Qwen3.6-35B-A3Bが強すぎる！！！

・opencode,vibe-local,GitHub Copilot,qwencode,claude codeと組み合わせたときのts-benchを実施したところ、すべて満点

・しかもClaude sonnet 4.6やOpus 4.6と同じくらい速くタスクを遂行できている

Qwen3.5-27Bもすごかったが、Qwen3.6-35B-A3Bは赤い彗星のごとく27Bよりも推論速度が3倍速いので、ベンチマーク結果からもわかるようにタスク遂行までの時間が大幅に短縮できるようになったのが大きい

[Md Ismail Šojal 🕷️ @0x0SojalSec]

A fully local 26B MoE model was built for red teaming and bug hunting.

Trained on elite bug reports and real evasion tactics. DPO fine-tuned for hunter mindset.

Claude sees your payloads in logs; that's why BugTraceAI Apex 26B local MoE for real red teamers.

- executes WAF bypasses with internal thinking blocks.
- It enforces deep internal reasoning before generating any output.
- Delivers production-grade WAF/EDR evasion with AES-256-GCM obfuscation.
- Zero refusals, Trained on real-world elite reports and evasion techniques.

Fits in 16.7GB. Runs on RTX 3060.

- https://t.co/FfnGgGXoQy

[The Hacker News @TheHackersNews]

⚠️ ALERT - CPUID’s site was compromised for ~19 hours, serving trojanized CPU-Z and HWMonitor installers.

Attackers used DLL sideloading to pair legit apps with a malicious file, deploying STX RAT.

150+ victims reported before detection.

🔗 Read → https://t.co/3Oshrvbawo

[The Hacker News @TheHackersNews]

⚠️ A compromised AI library exposed developer machines.

1,705 packages pulled infected LiteLLM versions, harvesting SSH keys and cloud creds from local systems via dependencies.

It worked because secrets sit in plaintext across files and tools.

🔗 How one dependency exposed thousands of environments → https://t.co/OKB9Rgfsiw

[보안프로젝트 @ngnicky]

Getting a Shell on the Tapo C260 Camera (CVE-2026-0651, CVE-2026-0652, CVE-2026-0653) https://t.co/gjoRfoYws9

[Claude @claudeai]

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: https://t.co/n4SZ9EIklG