VXLANG is a project designed to prevent reverse-engineering behaviors such as static or dynamic analysis, file tampering, and unauthorized access to memory by attackers.
The vxlang project currently targets x86-64 system and .Net binaries, native binary files for the Microsoft Windows operating system, including executables with the ".exe" extension, dynamic link library files with the ".dll" extension, and kernel driver files with the ".sys" extension. (The types of target binaries supported by vxlang will be expanded in future updates).
Before
After
Before
After x64dbg
After IDA - TYPE 1
Optimization: Verify that VxVirtualizationBegin and VxVirtualizationEnd are in the same block.
If procedures are merged due to code optimization, problems may arise.switch-case: In the switch-case, when the branch movement is not an IMM value, it may be moved to the original position.Exception handling: Currently, the virtual CPU does not support exception handling.
The full version of VxLang is developed and operated under donation through Patreon. The exact usage of VxLang and sharing of extension module code, new features of VxLang, etc. will be shared. If you have any questions, please send your request to the email below.
- E-Mail: 0x999h@gmail.com
- Full Version
- You can control all features of the VxLang protector.
- Support for virtual CPUs with specialized CPU contexts.
- Patreon
1.0.3.0- [Obfuscator] Changed the code flattening difficulty via the SDK.
- [Packer] Fixed small bugs.
1.0.2.0- [Obfuscator] Fixed to make code flattening compile and behave faster.
- [Obfuscator] Fixed code that interferes with code analyzer.
- [Obfuscator] Fixed deadstore blocks.
- [Packer] Update packer shell code.
1.0.1.1- [Obfuscator] Fixed a critical bug in code flattening
- Fixed a bug where blocks would break if there were more than a certain amount of nodes
- [Virtualizer] Fixed a critical bug in code virtualizer
- Fixed a bug where virtual machine regions were calculated incorrectly, causing packers to fail
- [Packer] Add packer shell code
- Added shell code to slow down the code analyzer
- Virtualization and obfuscation rotation
- [Obfuscator] Fixed a critical bug in code flattening
1.0.1.0- [Obfuscator] Code flattening is now even faster.
- [Obfuscator] Code flattening blocks have increased significantly.
- [Obfuscator] Added dummy code to delay code analysis.
- [Virtualizer] Fixed a bug with --opt-ref-call.
- [Virtualizer] The virtual machine has been rotated.
- [Packer] Removed the compiler pattern for PE format.
1.0.0.5- Support code flattening for kernel drivers.
- Speeding up the code flattening process.
- Increased the difficulty of code flattening and code obfuscation.
- Added techniques like dead-stores.
- The virtual machine has been updated.
1.0.0.4- Code flattening is applied to the Virtual-Machine.
- Adjusts the difficulty of Code-Flattening.
- Virtual Machines and Code Obfuscation rotated.
- Fixed some shell code.
- Fixed a small bug.
1.0.0.3- Adjust the speed of code flattening (compiles faster)
- Randomly modify code flattening context (more updates coming soon)
- Fixed bug:https://www.patreon.com/posts/92864752?utm_campaign=postshare_creator
- Virtual Machine Rotation.
1.0.0.2- Adjust the packing speed.
- Increases the virtualization scope of shell code.
- The above changes continuously.
1.0.0.1- Code-flattening can now be applied to packer shell code as well.
- Fixed a small bug.
- Obfuscation has been rotated.
0.9.9.14- Encrypt resource data. (In a future update, all but static resource data will be hidden.)
- Fixed where to look for keys from the current directory to the VxLang location.
- Fixed a bug with log output.
- The number of virtual machines has been added to 70+.
- Obfuscation has been rotated.
0.9.9.13- Fixed the appearance of code flattening output
- Add Code Flattening API
- VL_CODE_FLATTENING_TYPE2_BEGIN / VL_CODE_FLATTENING_TYPE2_END
- Obfuscation rotation
- Fixed operating speed
- Fix SDK missing
0.9.9.12- Added code virtualization to the packer shell code
- The VxLang key has been updated
0.9.9.11- Fixed node depth for code flattening
- Rotated obfuscation
0.9.9.10- User-reported bug fixes
- Critical bug: Bug fix for import table protection
- Fixed a bug where some anti-debugging features were automatically enabled
- Thank you so much for reporting this :)
- User-reported bug fixes
0.9.9.9- Added
--opt-section-name, which sets the section name to be based on.- Ex) vxlang.exe ${EXE} --opt-section-name .test
- Rotated the license key and obfuscation.
- Fixed a few small bugs.
- Added
0.9.9.8- Fixed a bug where the
Flatten Entry Pointsoption was not being applied when run on its own.- You can run
--opt-fla-entryon its own, without adding any code flattening related SDKs.
- You can run
- Fixed a bug where the
0.9.9.7- Add 50 or more virtual machines.
- Rotate obfuscation.
0.9.9.6- Added key checking for VxLang: invalid keys will prevent execution.
- Fixed a conflict with some features in Axion mode.
- Fixed plugin arguments.
0.9.9.5- Remove debug information (PDB) by default.
- Adding
--use-debug-datawill generate debug information.
- Adding
- Some of the packer shell code has become a code virtualization task.
- Remove debug information (PDB) by default.
0.9.9.4- Add 30+ virtual machines.
- Contexts/handlers/handle tables/link tables/etc. are all different.
- Modify the VxLang context for "Customizing the full-version of VxLang Part 3".
- Update download links.
0.9.9.3- Added commands for virtual machines.
- Fixed a small bug.
0.9.9.2- Fixed bugs with protector.
- Fixed the Protector API.
0.9.9.1- Updated the core(Virtual-CPU) of code virtualization.
- All handler and link code, and handle tables are randomized.
- Updated the core(Virtual-CPU) of code virtualization.
1.0.0- develop the ELF editor.
- test and bug fix for Windows version.
Task- Add ELF32/64 format for x86-64
- Add code obfuscation methods for x86-64
- Once the above steps are complete, add the ARM system for the ELF file format.
Thank you to everyone who helped with the development.