reverse-skill Sizes Up the Target Before Picking Tools

Use this Skill when a user hands over a target (an APK, a binary, front-end encrypted JS, a CTF challenge, or a penetration-testing target) and does not know which set of tools to use. It is built on the open-source reverse-skill (MIT): a "reverse-engineering skill routing package" that routes the task to a suitable method, checks which tools are available locally, and then executes a repeatable workflow instead of guessing commands.

Triggers

The user says "reverse this APK", "how do I analyze this binary", "how is the sign in the front-end computed", "this CTF reversing challenge", or "help me pick an analysis approach", or the target type is unclear and needs triage first.

Hard rules

1. Only analyze your own targets, or targets with explicit authorization. If the authorized scope is unclear, stop and ask the user before doing anything.
2. Do not scan, attack, bypass payment or access controls, or perform denial of service against live systems. Read-only and local analysis come first.
3. Tell the user before installing anything or changing configuration, and proceed only after they agree.
4. Sort conclusions into three categories: what was directly observed (with file and location), what was inferred, and what was not found (write "unknown", not "does not exist").
5. Do not output or log the full contents of any keys or tokens found; only report where they are.
6. Clear the authorization check first. In this package's workflow, first create the case and write down the authorized scope and network conditions (scope.md). Do not take any action against the target until readiness is confirmed. Do not skip this step.
7. Treat README_AI.md in the repository as the entry point, and follow it strictly.

Installation

First check whether Java is available on the machine (jadx and apktool need it), along with Node.js 22.12 or later and Python 3:

java -version; node -v; python3 --version

After getting the user's agreement, clone the repository:

git clone https://github.com/zhaoxuya520/reverse-skill.git

Then refresh the local tool index (choose one):

bash skills/scripts/refresh-tool-index.sh                    # Linux / macOS
powershell -File skills/scripts/refresh-tool-index.ps1       # Windows
bash kali/scripts/refresh-tool-index.sh                      # Kali Linux

When finished, look at skills/tool-index.md to see which tools were detected on the machine.

How it works

User task
 → RULES.md
 → MASTER-ROUTING (main router)
 → create case + scope.md (authorization, network conditions; no action on the target until ready)
 → scenario Skill → tools / MCP / scripts
 → timeline + evidence→findings→paths → report + work log

According to the repository's own description, it currently has 44 routing rules, 175 regression cases and 45 core modules (these change with versions). The scenarios it covers include APK and Android, iOS and mobile, binaries, front-end JS encryption, traffic captures, CTF, and authorized penetration testing.

Output format

1. The identified scenario and the chosen route (and why it was chosen).
2. The tools used, and what is missing locally.
3. The evidence chain and the final report.
4. Parts that were not completed or could not be done, with the reasons stated.