Analysis updated 2026-07-17 · repo last pushed 2026-07-03
Route an AI coding assistant through APK analysis, JavaScript deobfuscation, and HTTP capture to understand an Android app's network encryption.
Use the CTF module's 40+ sub-skills to guide an AI through competition reverse-engineering challenges.
Get structured binary analysis workflows using tools like IDA Pro and radare2 during authorized penetration tests.
Analyze firmware or IoT devices with a routing framework instead of manually picking tools each time.
| zhaoxuya520/reverse-skill | jenkinsci/docker | i-am-jakoby/flipper-zero-badusb | |
|---|---|---|---|
| Stars | 7,205 | 7,491 | 6,809 |
| Language | PowerShell | PowerShell | PowerShell |
| Last pushed | 2026-07-03 | — | — |
| Maintenance | Active | — | — |
| Setup difficulty | moderate | moderate | moderate |
| Complexity | 3/5 | 3/5 | 2/5 |
| Audience | developer | ops devops | researcher |
Figures from each repo's GitHub metadata at analysis time.
Not a standalone app, install alongside an AI coding client and requires Java, Node.js, and Python already on your machine.
reverse-skill is a toolkit that makes AI coding assistants like Claude Code, Cursor, or Codex CLI genuinely useful for security work, reverse engineering Android apps, analyzing malware, penetration testing, solving CTF challenges, and similar tasks. Normally, if you ask an AI agent to dissect a compiled binary or find a vulnerability, it guesses at commands and often gets things wrong. This project gives the AI a structured decision-making framework so it picks the right tool and methodology for each specific task. The core idea is a routing system. When you give the AI a task, say, unpacking an APK or deobfuscating JavaScript, a set of rules (laid out in RULES.md) guides it through a routing matrix that matches the task to the correct skill module. Each skill module knows which tools to run and how. So instead of the AI blindly trying random commands, it follows a proven workflow: identify what kind of file or problem you're dealing with, route to the matching skill, execute the right tools in sequence, then produce a report. It also saves what it learned, so future tasks benefit from past experience. The practical users are security researchers, penetration testers, CTF competitors, and anyone doing authorized security analysis. If you're trying to understand how an Android app encrypts its network traffic, this system would route the AI through APK analysis tools, then JavaScript reverse engineering, then HTTP capture, in the right order, with the right tools at each step. It covers a broad range: binary analysis with tools like IDA Pro and radare2, firmware and IoT testing, vulnerability exploitation, EDR bypass research, and even LLM/AI security testing. The CTF module alone bundles over 40 sub-skills for competition scenarios. The project is built as a collection of markdown files and scripts rather than a traditional application. You don't really "run" it directly, you install it alongside your AI coding client, and the AI reads the instructions to understand what tools are available and how to approach different problems. It supports Linux, macOS, and Kali (a security-focused Linux distribution), and relies on existing tools like Java, Node.js, and Python being installed on your machine.
Toolkit that gives AI coding assistants a structured routing system for security work like Android reverse engineering, malware analysis, penetration testing, and CTF challenges.
Mainly PowerShell. The stack also includes PowerShell, Java, Node.js.
Active — commit in last 30 days (last push 2026-07-03).
License is not stated in the available content.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.