zhaoxuya520/reverse-skill

What it solves

AI agents often struggle with cybersecurity tasks because they don't know which specific tool (like jadx, Frida, or IDA Pro) to use for a particular file type or scenario, leading to guesswork and repeated mistakes. This project provides a structured routing system that guides AI agents through the correct methodology and toolchain for reverse engineering and penetration testing.

How it works

The project acts as a "skills router" for AI clients (such as Claude Code, Cursor, or Codex). It uses a central configuration file (routing.json) and a set of routing rules to map a user's task to a specific security scenario. Once a scenario is identified, the agent is routed to a repeatable workflow involving specific scripts, MCP servers, and tools. The process follows a strict pipeline: task analysis $\rightarrow$ routing $\rightarrow$ case initialization $\rightarrow$ scenario-specific skill execution $\rightarrow$ evidence collection $\rightarrow$ reporting.

Who it’s for

Security researchers, CTF competitors, and penetration testers who use AI agents to automate or assist in reverse engineering, malware analysis, and vulnerability research.

Highlights

• Comprehensive Scenario Coverage: Supports a wide array of tasks including APK/iOS analysis, binary reverse engineering, JS encryption, firmware pentesting, and LLM security.
• Client-Neutral Core: Designed to work across different AI clients without being locked into a single provider.
• Automated Tool Indexing: Includes scripts to detect and index available local tools across Windows, Ubuntu, and macOS.
• Rigorous Validation: Features a regression benchmark with 175 cases to ensure routing accuracy and CI pipelines to verify structural coherence.