agentleFS
mohitmishra786mohitmishra786/low-level-dev-skills/skills/security/reverse-engineering/SKILL.md

reverse-engineering

Skill290 starsChanged 7 months agoMITView on GitHub

Reverse engineering skill for binary analysis. Use when decompiling with Ghidra, analyzing with radare2, scripting RE tools, triaging with strings/file/xxd, or diffing binaries. Activates on queries about Ghidra, radare2, r2, decompiler, Binary Ninja, Diaphora, or stripped binary analysis.

34 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use ida-reverse, game-automation and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • DorianGalloSkill

    ida-reverse

    IDA Pro 逆向分析辅助技能。当用户提到逆向、反编译、分析二进制/PE/ELF/APK/DLL/SO、破解、找密码、漏洞分析、病毒分析、firmware 固件分析,或需要分析 exe/dll/so/elf/macho/sys 等文件时,务必使用此技能。

    Also in 6 projects that use this

  • rehan-remadeSkill

    game-automation

    Launch, see and drive a real game so an agent can test its own mods. Covers launching via Steam or the exe, windowed mode at a fixed size, screenshots of the game window (GPU-safe capture), clicking and typing into menus, holding keys, relative mouse, crash-reporter cleanup, killing by PID, reading logs, scripted test scenes, and an in-game JSON bridge that lets an agent observe and act. Windows-first (also from WSL), with notes for Linux and macOS. Use when a mod must be verified in game, when navigating menus automatically, or when building a repeatable test or demo scene.

    Also in 4 projects that use this

  • rehan-remadeSkill

    game-recon

    Figure out how a specific installed game can be modded, before writing any mod code. Covers where it's installed, the engine and version, managed vs native code, anti-cheat, loaders already present, save and config folders, and the community's modding route. Also whether this machine can run it (a Windows game on macOS or Linux, x64 on ARM). Use when the user asks "can I mod <game>?", "what engine is <game>", "how do people mod <game>", or at the start of any modding task. Produces MODDING_PLAN.md.

    Also in 4 projects that use this

  • DorianGalloSkill

    attack-chain

    Also in 5 projects that use this

  • +4 more from rehan-remade
  • +4 more from DorianGallo

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 27 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use diffing to find patched vulnerability functions after updates.

    line 179

“Common” is measured: kept by at least 6 in 10 of the 27 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Reverse Engineering

Purpose

Guide agents through reverse engineering binaries: Ghidra project setup and decompilation, radare2 analysis workflow, Binary Ninja scripting, initial triage with strings/file/xxd, identifying C++ patterns (vtables, RAII), analyzing stripped binaries, and diffing with Diaphora or BinDiff.

When to Use

  • Analyzing an unknown binary without source code
  • Recovering algorithm logic from compiled executables
  • Comparing two firmware versions for vulnerability patches
  • Understanding malware or CTF challenge binaries
  • Recovering symbols from stripped ELF/PE files
  • Automating analysis with Ghidra or r2 scripts

Workflow

1. Initial triage

file suspicious_binary
strings -n 8 suspicious_binary | head -50
strings -el suspicious_binary          # UTF-16 LE
xxd suspicious_binary | head -20
readelf -h suspicious_binary           # ELF
objdump -d -M intel suspicious_binary | head -40

# Check protections
checksec --file=suspicious_binary
CommandReveals
fileArchitecture, static/dynamic, stripped
stringsURLs, paths, error messages, keys
readelf -sSymbol table (if not stripped)
nm -DDynamic symbols
checksecRELRO, NX, PIE, canary

2. Ghidra workflow

# Headless analysis
analyzeHeadless /tmp/ghidra_projects MyProject \
  -import suspicious_binary \
  -postScript ExportDecompile.java

# GUI: File → New Project → Import File → Analyze (Yes)

Key steps:

  1. Auto-analysis — let Ghidra complete disassembly
  2. Define functions — F at entry points if missed
  3. Decompiler — Window → Decompiler (C-like output)
  4. Rename — L on variables/functions for clarity
  5. Cross-references — Ctrl+Shift+F on function/data
// Ghidra script (Java) — list functions > 100 bytes
import ghidra.program.model.listing.*;

FunctionManager fm = currentProgram.getFunctionManager();
for (Function f : fm.getFunctions(true)) {
    if (f.getBody().getNumAddresses() > 100)
        println(f.getName() + " @ " + f.getEntryPoint());
}
# Ghidra Python (Jython)
from ghidra.program.model.listing import FunctionManager
fm = currentProgram.getFunctionManager()
for f in fm.getFunctions(True):
    print(f.getName(), f.getEntryPoint())

3. radare2 workflow

r2 suspicious_binary
[0x00001000]> aaa          # analyze all
[0x00001000]> afl          # list functions
[0x00001000]> pdf @ main   # disassemble function
[0x00001000]> VV           # visual graph mode
[0x00001000]> iz           # strings in data sections
[0x00001000]> s sym.main; pdf

Patching:

[0x00001000]> wx 9090 @ 0x401234   # write NOPs
[0x00001000]> wci 0x401234         # insert instruction
[0x00001000]> wt modified_binary
# r2 scripting
r2 -qc 'aaa; afl' suspicious_binary
r2 -i analysis.r2 suspicious_binary

4. Binary Ninja scripting

# BN Python API
import binaryninja as bn

bv = bn.load("suspicious_binary")
for func in bv.functions:
    if func.name.startswith("sub_"):
        hlil = func.hlil
        for block in hlil:
            print(block)

5. C++ pattern recognition

// Vtable pattern in disassembly
// mov rax, [rdi]      ; load vtable pointer
// call [rax+0x10]     ; virtual call at offset

// Constructor pattern
// mov [obj], offset vtable
PatternIndicator
Vtable.data.rel.ro section, array of function pointers
RAIIpaired ctor/dtor calls, exception landing pads
TemplatesMangled names _Z..., duplicate logic per type
std::stringSSO buffer inline or heap pointer at offset 0
# Demangle C++ symbols
c++filt _ZN4Math3addEii

6. Stripped binary recovery

# Find main via __libc_start_main
readelf -s binary | grep -E 'main|start'
# Or r2: afl~entry

# FLIRT signatures (Ghidra/BN) — match libc patterns
# Stack string analysis in Ghidra decompiler

Strategies:

  • Identify main via libc init or entry point
  • Find syscalls (syscall insn on Linux)
  • String xref to locate error handlers
  • Entropy analysis for encrypted sections

7. Binary diffing

# Diaphora (Ghidra/IDA plugin)
# Export from both binaries, run diff

# BinDiff (commercial, IDA/Ghidra)
bindiff old.i64 new.i64

# Simple hash diff
sha256sum firmware_v1 firmware_v2
diff <(objdump -d v1) <(objdump -d v2) | head

Use diffing to find patched vulnerability functions after updates.

8. RE decision tree

Binary type?
├── ELF/Linux → Ghidra + r2 + readelf
├── PE/Windows → Ghidra + PE-bear + x64dbg reference
├── Firmware → binwalk extract → Ghidra on architecture
└── Obfuscated → dynamic analysis (gdb/ltrace) first

Common Problems

SymptomCauseFix
Ghidra decompiler failsIndirect jumps, bad typesFix function signature; define struct
r2 analysis incompleteLarge binaryaaa then aac ; increase analysis depth
Wrong architectureARM vs Thumb, MIPSSet -a arm or correct Ghidra language
Anti-debug trapptrace checkPatch or use -gdb in QEMU
Packed binaryUPX/etc.upx -d or manual unpack
No xrefs to stringPIE/RELROFollow GOT; runtime analysis

Related Skills

  • skills/binaries/elf-inspection — ELF structure analysis
  • skills/debuggers/gdb — dynamic analysis complement
  • skills/runtimes/binary-hardening — understanding mitigations being bypassed
  • skills/security/kernel-security — kernel RE and CVE analysis
  • skills/low-level-programming/assembly-x86 — reading disassembly
  • skills/low-level-programming/assembly-arm — ARM binary analysis

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.