Python Obfuscator is a tool to obfuscate, virtualize & protect Python .py scripts against reverse engineering with a VM engine (or finite-state automata machine), self-defending integrity checks, polymorphic string encryption, anti-debugging, anti-vm, anti-emulation and 50+ total obfuscation strategies!
It's available for Windows & Linux, plus VSCode/Cursor extension:
Multiple programming APIs available:
An online obfuscator interface:
Scripts are typically distributed as plain .py files or bundled inside packages. That convenience means anyone with file access can read the full logic, hunt for credentials or API keys in strings, and steal your algorithms unless you take extra steps to hide intent.
Python is one of the most popular general-purpose scripting languages. It is widely used for automation, backend services, data processing, tooling, and licensing/agent code that customers run on their own machines.
Python Obfuscator comes with many advanced obfuscation, virtualization & protection strategies. You can easily tune protection versus size and performance.
Polymorphic string encryption is the core of the string pipeline: each build generates a fresh decryptor so literals never share one static algorithm. Integers, floats, and decoy noise join that layer. The result conceals literals and structure while preserving tested runtime behaviour.
Code virtualization is the main control-flow defense. Selected statements are lifted into a randomly generated VM engine (or FSA / flattened dispatcher) with shuffled dispatch tables, decoy opcodes, and an obfuscated dispatcher loop. Analysts must interpret the virtual machine instead of reading plain Python.
Finite-state automata (FSA) obfuscation rewrites linear Python statement blocks into dual-state automata with opaque schedulers and shuffled dispatch handlers. Instead of reading code top to bottom, analysts must follow numeric states, transition tables, and decoy paths to reconstruct the original order.
Anti-debugging protection inserts polymorphic probes that detect attached debuggers, tracing hooks, and related host signals, together with anti-VM, anti-sandbox and anti-emulator checks. When a check fires, the obfuscated script exits silently instead of revealing protected logic under interactive analysis.
Self-defending integrity checks verify that the obfuscated file has not been patched. A bootstrap probe hashes the on-disk script and sets a tamper key when it no longer matches the obfuscated build. String decryptors consume that key, so edited scripts return garbage instead of plaintext.
The protection linker adds decoy functions and fake calls so a copied fragment still looks like real program code. Hidden traps fire only if someone edits the file or runs a piece of it on its own. When the script starts normally, those extras stay silent and the program runs as usual.
Look at this example — the same script becomes harder to read at a glance after obfuscation.
label = 'SecretKey'
port = 443
def get_sum(a, b):
return a + b
print(label, port)
r = get_sum(11, 31)
print(r)
Would you still recognise the original intent if you only had the obfuscated text and no prior copy of the script?
The engine parses Python source into an AST tree, then applies selectable transforms. Code virtualization (VM, FSA, or flattening) rewrites execution so analysts cannot read the script top to bottom. Polymorphic string encryption hides literals; self-defending integrity probes detect patched files. Also available: identifier renaming, numeric encryption, noise and decoy insertion, the protection linker, and anti-debugging checks. Many techniques are specific to this product; some ideas are shared with our other protection tools.
When all passes finish, the engine emits a new .py file. Edge cases in the Python grammar and hosting environments mean you should always test the output in your target runtime.
Take no chances, use Python Obfuscator to obfuscate, virtualize and protect your Python scripts and algorithms.
Our company has a long history in obfuscation technologies and code obfuscators (see our PowerShell, Java, AutoIt, x86 Assembly & obfuscators).
We actively bugfix, research and develop new obfuscation strategies for our tools.
You can count on our expertise and support in this field.
The preferred way to install the Web API SDK is via RubyGems.
gem install python-obfuscatorrequire "python-obfuscator"The gem is published at https://rubygems.org/gems/python-obfuscator
# frozen_string_literal: true
###############################################################################
#
# Python Obfuscator — login
#
# Version : v1.0.0
# Language : Ruby
# Author : Bartosz Wójcik
# Web page : https://www.pelock.com
#
###############################################################################
$LOAD_PATH.unshift(File.expand_path("../lib", __dir__))
require "python-obfuscator"
p PythonObfuscator.new("YOUR-WEB-API-KEY").login# frozen_string_literal: true
###############################################################################
#
# Python Obfuscator — default options
#
# Version : v1.0.0
# Language : Ruby
# Author : Bartosz Wójcik
# Web page : https://www.pelock.com
#
###############################################################################
$LOAD_PATH.unshift(File.expand_path("../lib", __dir__))
require "python-obfuscator"
client = PythonObfuscator.new("YOUR-WEB-API-KEY")
source = <<~PY
label = 'SecretKey'
port = 443
def get_sum(a, b):
return a + b
print(label, port)
r = get_sum(11, 31)
print(r)
PY
result = client.obfuscate_script_source(source)
if result && result["error"] == PythonObfuscator::ERROR_SUCCESS
puts result["output"]
else
warn "Error: #{result.inspect}"
endSee the examples/ directory in this repository for complete samples.
Online interface for Python Obfuscator is available at:
https://www.pelock.com/python-obfuscator/
You can download it at:
https://www.pelock.com/products/python-obfuscator/download
Python Obfuscator ships with a command-line interface for Windows and Linux automation. Use it to integrate obfuscation into build servers, CI jobs, or batch packaging.
In demo mode the obfuscator always applies integers_to_arrays, mba_binops, and encrypt_strings (no code virtualization). Source size is limited to 1000 characters.
Bartosz Wójcik
- Visit my site at — https://www.pelock.com
- X — https://x.com/PELock
- GitHub — https://github.com/PELock