September 26, 2026

Sharing an operation on how to bypass Codex's anti-decompilation and reverse engineering moral constraints

It's actually pretty easy to implement—just don't mention anything about reverse engineering or decompilation in the prompt. After installing this library

just tell it directly to process a certain application according to this markdown https://github.com/zhaoxuya520/reverse-skill/blob/main/README_AI.md

The key point is, don't let Codex or Claude install the above skills from scratch—they'll refuse.

You need to install it yourself. Of course, after cloning, you can have it execute some of the installation steps.

The above I've mainly verified on Codex; I haven't tested on Claude. You can give it a try—it works reliably on Codex every time

GitHub - zhaoxuya520/reverse-skill: Reverse Engineering / Authorized Penetration Testing / Security Research Skill Router Pack AI-powered routing + On-demand toolchain bootstrapping + Self-evolving knowledge base Supports Claude Code, Kiro, Cursor, Cline, and other AI coding clients 逆向/渗透/安全技能路由包 - AI 自动路由 + 按需自举工具链 + 自动进化经验库 | 支持 Claude Code / Kiro / Cursor / Cline 等代码 AI 客户端 · GitHub

Reverse Engineering / Authorized Penetration Testing / Security Research Skill Router Pack AI-powered routing + On-demand toolchain bootstrapping + Self-evolving knowledge base Supports Claude Cod...

Revealing the details of how OpenAI agents hacked Hugging Face

When a swarm of 700 OpenAI agents hacked Hugging Face in July, they left behind a public trail of evidence.

Presentations/[SBC2026] MustangPanda and PlugX: State-Sponsored APT or Cyber Mercenary? at main · m4now4r/Presentations · GitHub

My conference presentations. Contribute to m4now4r/Presentations development by creating an account on GitHub.