AGENT SKILL

anti-reversing-techniques

Understand anti-reversing, obfuscation, and protection techniques encountered during software analysis. Use when analyzing protected binaries, bypassing anti-debugging for authorized analysis, or understanding software protection mechanisms.

License 许可
MIT
最近更新
2026-11-10
Stars
★ 1
Forks
0

01 前言 Front Matter

name
anti-reversing-techniques
description
Understand anti-reversing, obfuscation, and protection techniques encountered during software analysis. Use when analyzing protected binaries, bypassing anti-debugging for authorized analysis, or understanding software protection mechanisms.

03 SKILL.md 正文

AUTHORIZED USE ONLY: This skill contains dual-use security techniques. Before proceeding with any bypass or analysis:

  1. Verify authorization: Confirm you have explicit written permission from the software owner, or are operating within a legitimate security context (CTF, authorized pentest, malware analysis, security research)
  2. Document scope: Ensure your activities fall within the defined scope of your authorization
  3. Legal compliance: Understand that unauthorized bypassing of software protection may violate laws (CFAA, DMCA anti-circumvention, etc.)

Legitimate use cases: Malware analysis, authorized penetration testing, CTF competitions, academic security research, analyzing software you own/have rights to

Anti-Reversing Techniques

Understanding protection mechanisms encountered during authorized software analysis, security research, and malware analysis. This knowledge helps analysts bypass protections to complete legitimate analysis tasks.

Anti-Debugging Techniques

Windows Anti-Debugging

API-Based Detection

// IsDebuggerPresent
if (IsDebuggerPresent()) {
    exit(1);
}

// CheckRemoteDebuggerPresent
BOOL debugged = FALSE;
CheckRemoteDebuggerPresent(GetCurrentProcess(), &debugged);
if (debugged) exit(1);

// NtQueryInformationProcess
typedef NTSTATUS (NTAPI *pNtQueryInformationProcess)(
    HANDLE, PROCESSINFOCLASS, PVOID, ULONG, PULONG);

DWORD debugPort = 0;
NtQueryInformationProcess(
    GetCurrentProcess(),
    ProcessDebugPort,        // 7
    &debugPort,
    sizeof(debugPort),
    NULL
);
if (debugPort != 0) exit(1);

// Debug flags
DWORD debugFlags = 0;
NtQueryInformationProcess(
    GetCurrentProcess(),
    ProcessDebugFlags,       // 0x1F
    &debugFlags,
    sizeof(debugFlags),
    NULL
);
if (debugFlags == 0) exit(1);  // 0 means being debugged

Bypass Approaches:

# x64dbg: ScyllaHide plugin
# Patches common anti-debug checks

# Manual patching in debugger:
# - Set IsDebuggerPresent return to 0
# - Patch PEB.BeingDebugged to 0
# - Hook NtQueryInformationProcess

# IDAPython: Patch checks
ida_bytes.patch_byte(check_addr, 0x90)  # NOP

PEB-Based Detection

// Direct PEB access
#ifdef _WIN64
    PPEB peb = (PPEB)__readgsqword(0x60);
#else
    PPEB peb = (PPEB)__readfsdword(0x30);
#endif

// BeingDebugged flag
if (peb->BeingDebugged) exit(1);

// NtGlobalFlag
// Debugged: 0x70 (FLG_HEAP_ENABLE_TAIL_CHECK |
//                 FLG_HEAP_ENABLE_FREE_CHECK |
//                 FLG_HEAP_VALIDATE_PARAMETERS)
if (peb->NtGlobalFlag & 0x70) exit(1);

// Heap flags
PDWORD heapFlags = (PDWORD)((PBYTE)peb->ProcessHeap + 0x70);
if (*heapFlags & 0x50000062) exit(1);

Bypass Approaches:

; In debugger, modify PEB directly
; x64dbg: dump at gs:[60] (x64) or fs:[30] (x86)
; Set BeingDebugged (offset 2) to 0
; Clear NtGlobalFlag (offset 0xBC for x64)

Timing-Based Detection

// RDTSC timing
uint64_t start = __rdtsc();
// ... some code ...
uint64_t end = __rdtsc();
if ((end - start) > THRESHOLD) exit(1);

// QueryPerformanceCounter
LARGE_INTEGER start, end, freq;
QueryPerformanceFrequency(&freq);
QueryPerformanceCounter(&start);
// ... code ...
QueryPerformanceCounter(&end);
double elapsed = (double)(end.QuadPart - start.QuadPart) / freq.QuadPart;
if (elapsed > 0.1) exit(1);  // Too slow = debugger

// GetTickCount
DWORD start = GetTickCount();
// ... code ...
if (GetTickCount() - start > 1000) exit(1);

Bypass Approaches:

- Use hardware breakpoints instead of software
- Patch timing checks
- Use VM with controlled time
- Hook timing APIs to return consistent values

Exception-Based Detection

// SEH-based detection
__try {
    __asm { int 3 }  // Software breakpoint
}
__except(EXCEPTION_EXECUTE_HANDLER) {
    // Normal execution: exception caught
    return;
}
// Debugger ate the exception
exit(1);

// VEH-based detection
LONG CALLBACK VectoredHandler(PEXCEPTION_POINTERS ep) {
    if (ep->ExceptionRecord->ExceptionCode == EXCEPTION_BREAKPOINT) {
        ep->ContextRecord->Rip++;  // Skip INT3
        return EXCEPTION_CONTINUE_EXECUTION;
    }
    return EXCEPTION_CONTINUE_SEARCH;
}

Linux Anti-Debugging

// ptrace self-trace
if (ptrace(PTRACE_TRACEME, 0, NULL, NULL) == -1) {
    // Already being traced
    exit(1);
}

// /proc/self/status
FILE *f = fopen("/proc/self/status", "r");
char line[256];
while (fgets(line, sizeof(line), f)) {
    if (strncmp(line, "TracerPid:", 10) == 0) {
        int tracer_pid = atoi(line + 10);
        if (tracer_pid != 0) exit(1);
    }
}

// Parent process check
if (getppid() != 1 && strcmp(get_process_name(getppid()), "bash") != 0) {
    // Unusual parent (might be debugger)
}

Bypass Approaches:

# LD_PRELOAD to hook ptrace
# Compile: gcc -shared -fPIC -o hook.so hook.c
long ptrace(int request, ...) {
    return 0;  // Always succeed
}

# Usage
LD_PRELOAD=./hook.so ./target

Anti-VM Detection

Hardware Fingerprinting

// CPUID-based detection
int cpuid_info[4];
__cpuid(cpuid_info, 1);
// Check hypervisor bit (bit 31 of ECX)
if (cpuid_info[2] & (1 << 31)) {
    // Running in hypervisor
}

// CPUID brand string
__cpuid(cpuid_info, 0x40000000);
char vendor[13] = {0};
memcpy(vendor, &cpuid_info[1], 12);
// "VMwareVMware", "Microsoft Hv", "KVMKVMKVM", "VBoxVBoxVBox"

// MAC address prefix
// VMware: 00:0C:29, 00:50:56
// VirtualBox: 08:00:27
// Hyper-V: 00:15:5D

Registry/File Detection

// Windows registry keys
// HKLM\SOFTWARE\VMware, Inc.\VMware Tools
// HKLM\SOFTWARE\Oracle\VirtualBox Guest Additions
// HKLM\HARDWARE\ACPI\DSDT\VBOX__

// Files
// C:\Windows\System32\drivers\vmmouse.sys
// C:\Windows\System32\drivers\vmhgfs.sys
// C:\Windows\System32\drivers\VBoxMouse.sys

// Processes
// vmtoolsd.exe, vmwaretray.exe
// VBoxService.exe, VBoxTray.exe

Timing-Based VM Detection

// VM exits cause timing anomalies
uint64_t start = __rdtsc();
__cpuid(cpuid_info, 0);  // Causes VM exit
uint64_t end = __rdtsc();
if ((end - start) > 500) {
    // Likely in VM (CPUID takes longer)
}

Bypass Approaches:

- Use bare-metal analysis environment
- Harden VM (remove guest tools, change MAC)
- Patch detection code
- Use specialized analysis VMs (FLARE-VM)

Code Obfuscation

Control Flow Obfuscation

Control Flow Flattening

// Original
if (cond) {
    func_a();
} else {
    func_b();
}
func_c();

// Flattened
int state = 0;
while (1) {
    switch (state) {
        case 0:
            state = cond ? 1 : 2;
            break;
        case 1:
            func_a();
            state = 3;
            break;
        case 2:
            func_b();
            state = 3;
            break;
        case 3:
            func_c();
            return;
    }
}

Analysis Approach:

  • Identify state variable
  • Map state transitions
  • Reconstruct original flow
  • Tools: D-810 (IDA), SATURN

Opaque Predicates

// Always true, but complex to analyze
int x = rand();
if ((x * x) >= 0) {  // Always true
    real_code();
} else {
    junk_code();  // Dead code
}

// Always false
if ((x * (x + 1)) % 2 == 1) {  // Product of consecutive = even
    junk_code();
}

Analysis Approach:

  • Identify constant expressions
  • Symbolic execution to prove predicates
  • Pattern matching for known opaque predicates

Data Obfuscation

String Encryption

// XOR encryption
char decrypt_string(char *enc, int len, char key) {
    char *dec = malloc(len + 1);
    for (int i = 0; i < len; i++) {
        dec[i] = enc[i] ^ key;
    }
    dec[len] = 0;
    return dec;
}

// Stack strings
char url[20];
url[0] = 'h'; url[1] = 't'; url[2] = 't'; url[3] = 'p';
url[4] = ':'; url[5] = '/'; url[6] = '/';
// ...

Analysis Approach:

# FLOSS for automatic string deobfuscation
floss malware.exe

# IDAPython string decryption
def decrypt_xor(ea, length, key):
    result = ""
    for i in range(length):
        byte = ida_bytes.get_byte(ea + i)
        result += chr(byte ^ key)
    return result

API Obfuscation

// Dynamic API resolution
typedef HANDLE (WINAPI *pCreateFileW)(LPCWSTR, DWORD, DWORD,
    LPSECURITY_ATTRIBUTES, DWORD, DWORD, HANDLE);

HMODULE kernel32 = LoadLibraryA("kernel32.dll");
pCreateFileW myCreateFile = (pCreateFileW)GetProcAddress(
    kernel32, "CreateFileW");

// API hashing
DWORD hash_api(char *name) {
    DWORD hash = 0;
    while (*name) {
        hash = ((hash >> 13) | (hash << 19)) + *name++;
    }
    return hash;
}
// Resolve by hash comparison instead of string

Analysis Approach:

  • Identify hash algorithm
  • Build hash database of known APIs
  • Use HashDB plugin for IDA
  • Dynamic analysis to resolve at runtime

Instruction-Level Obfuscation

Dead Code Insertion

; Original
mov eax, 1

; With dead code
push ebx           ; Dead
mov eax, 1
pop ebx            ; Dead
xor ecx, ecx       ; Dead
add ecx, ecx       ; Dead

Instruction Substitution

; Original: xor eax, eax (set to 0)
; Substitutions:
sub eax, eax
mov eax, 0
and eax, 0
lea eax, [0]

; Original: mov eax, 1
; Substitutions:
xor eax, eax
inc eax

push 1
pop eax

Packing and Encryption

Common Packers

UPX          - Open source, easy to unpack
Themida      - Commercial, VM-based protection
VMProtect    - Commercial, code virtualization
ASPack       - Compression packer
PECompact    - Compression packer
Enigma       - Commercial protector

Unpacking Methodology

1. Identify packer (DIE, Exeinfo PE, PEiD)

2. Static unpacking (if known packer):
   - UPX: upx -d packed.exe
   - Use existing unpackers

3. Dynamic unpacking:
   a. Find Original Entry Point (OEP)
   b. Set breakpoint on OEP
   c. Dump memory when OEP reached
   d. Fix import table (Scylla, ImpREC)

4. OEP finding techniques:
   - Hardware breakpoint on stack (ESP trick)
   - Break on common API calls (GetCommandLineA)
   - Trace and look for typical entry patterns

Manual Unpacking Example

1. Load packed binary in x64dbg
2. Note entry point (packer stub)
3. Use ESP trick:
   - Run to entry
   - Set hardware breakpoint on [ESP]
   - Run until breakpoint hits (after PUSHAD/POPAD)
4. Look for JMP to OEP
5. At OEP, use Scylla to:
   - Dump process
   - Find imports (IAT autosearch)
   - Fix dump

Virtualization-Based Protection

Code Virtualization

Original x86 code is converted to custom bytecode
interpreted by embedded VM at runtime.

Original:     VM Protected:
mov eax, 1    push vm_context
add eax, 2    call vm_entry
              ; VM interprets bytecode
              ; equivalent to original

Analysis Approaches

1. Identify VM components:
   - VM entry (dispatcher)
   - Handler table
   - Bytecode location
   - Virtual registers/stack

2. Trace execution:
   - Log handler calls
   - Map bytecode to operations
   - Understand instruction set

3. Lifting/devirtualization:
   - Map VM instructions back to native
   - Tools: VMAttack, SATURN, NoVmp

4. Symbolic execution:
   - Analyze VM semantically
   - angr, Triton

Bypass Strategies Summary

General Principles

  1. Understand the protection: Identify what technique is used
  2. Find the check: Locate protection code in binary
  3. Patch or hook: Modify check to always pass
  4. Use appropriate tools: ScyllaHide, x64dbg plugins
  5. Document findings: Keep notes on bypassed protections

Tool Recommendations

Anti-debug bypass:    ScyllaHide, TitanHide
Unpacking:           x64dbg + Scylla, OllyDumpEx
Deobfuscation:       D-810, SATURN, miasm
VM analysis:         VMAttack, NoVmp, manual tracing
String decryption:   FLOSS, custom scripts
Symbolic execution:  angr, Triton

Ethical Considerations

This knowledge should only be used for:

  • Authorized security research
  • Malware analysis (defensive)
  • CTF competitions
  • Understanding protections for legitimate purposes
  • Educational purposes

Never use to bypass protections for:

  • Software piracy
  • Unauthorized access
  • Malicious purposes

04 同仓库更多 Skills

YYC-Cube
YYC-Cube/YYC3-Knowledge-Base

vercel-sandbox

Run agent-browser + Chrome inside Vercel Sandbox microVMs for browser automation from any Vercel-deployed app. Use when the user needs browser automation in a Vercel app (Next.js, SvelteKit, Nuxt, Remix, Astro, etc.), wants to run headless Chrome without binary size limits, needs persistent browser sessions across commands, or wants ephemeral isolated browser environments. Triggers include "Vercel Sandbox browser", "microVM Chrome", "agent-browser in sandbox", "browser automation on Vercel", or any task requiring Chrome in a Vercel Sandbox.

★ 1
YYC-Cube
YYC-Cube/YYC3-Knowledge-Base

slack

Interact with Slack workspaces using browser automation. Use when the user needs to check unread channels, navigate Slack, send messages, extract data, find information, search conversations, or automate any Slack task. Triggers include "check my Slack", "what channels have unreads", "send a message to", "search Slack for", "extract from Slack", "find who said", or any task requiring programmatic Slack interaction.

★ 1
YYC-Cube
YYC-Cube/YYC3-Knowledge-Base

electron

Automate Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify, etc.) using agent-browser via Chrome DevTools Protocol. Use when the user needs to interact with an Electron app, automate a desktop app, connect to a running app, control a native app, or test an Electron application. Triggers include "automate Slack app", "control VS Code", "interact with Discord app", "test this Electron app", "connect to desktop app", or any task requiring automation of a native Electron application.

★ 1
YYC-Cube
YYC-Cube/YYC3-Knowledge-Base

dogfood

Systematically explore and test a web application to find bugs, UX issues, and other problems. Use when asked to "dogfood", "QA", "exploratory test", "find issues", "bug hunt", "test this app/site/platform", or review the quality of a web application. Produces a structured report with full reproduction evidence -- step-by-step screenshots, repro videos, and detailed repro steps for every issue -- so findings can be handed directly to the responsible teams.

★ 1
YYC-Cube
YYC-Cube/YYC3-Knowledge-Base

agent-browser

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

★ 1
YYC-Cube
YYC-Cube/YYC3-Knowledge-Base

shadcn

Manages shadcn components and projects — adding, searching, fixing, debugging, styling, and composing UI, including chat interfaces. Provides project context, component docs, and usage examples. Applies when working with shadcn/ui, component registries, presets, --preset codes, or any project with a components.json file. Also triggers for "shadcn init", "create an app with --preset", "switch to --preset", and for building a design system or applying a DESIGN.md or brand spec to shadcn/ui.

★ 1