Skip to content

Latest commit

 

History

History
80 lines (64 loc) · 5.02 KB

Features.md

File metadata and controls

80 lines (64 loc) · 5.02 KB

St3ix Obfuscator – Features

Current Features

Name Obfuscation

  • Class renaming – Renames classes to short or random identifiers (e.g. a, b, c or xk9m2p)
  • Method renaming – Obfuscate method names; handles override chains; excludes main, constructors, native, enum methods; optional homoglyph/invisible chars
  • Field renaming – Obfuscate field names; excludes serialVersionUID and enum constants; optional homoglyph/invisible chars
  • Configurable name length – Minimum length 1–32 characters (class, method, field)
  • Random vs sequential – Optional random names per build for stronger obfuscation
  • Exclude patterns – Skip classes/packages from renaming. Use * to exclude all classes; prefix matches (e.g. com.example excludes com.example.*). Built-in: java., javax., Bukkit, Minecraft, etc.
  • Package flattening – Inner classes get short names; outer package hierarchy preserved
  • Homoglyph obfuscation – Use Unicode lookalikes (e.g. Cyrillic а instead of Latin a) for class/method/field names. Sequential: fixed mapping; random: varies per char.
  • Invisible character injection – Zero-width chars (U+200B, etc.) in names. Safe for JVM; harder to detect and remove.

Data Obfuscation

  • Number obfuscation – Hides int constants via math expressions (123 → (50*3)-27, (a<<n)+b, etc.); decompilers show expressions, not literals. long/float/double still use XOR. Less prone to constant folding than simple XOR.
  • Array dimension obfuscation – Hides array sizes (new int[8] → new int[(8 ^ key) ^ key])
  • Boolean obfuscation – Hides true/false literals using XOR (ICONST_0/ICONST_1 → (value ^ key) ^ key)
  • String obfuscation – XOR encryption; inline decrypt at each use site (no central decoder). Key per class and per string/field. Static final String fields: ConstantValue removed; initialized in <clinit> with obfuscated code, so private static final String KEY = "secret" never appears as readable text. Strong against strings-tools and casual inspection.
  • Optional random keys – numberKeyRandom, arrayKeyRandom, booleanKeyRandom, stringKeyRandom for different keys/patterns per build
  • Debug info stripping – Removes source file names, line number table, and local variable table. Decompilers show var0, var1, etc., and lose source mappings. Configurable via debugInfoStrippingEnabled.

Configuration

  • YAML config – config.yml next to JAR
  • Per-feature toggles – Enable/disable class, method, field renaming; number, array, boolean, and string obfuscation; debug stripping
  • Random options – Toggle randomness for class/method/field names and obfuscation keys

Output & CLI

  • GUI – Graphical interface (run-gui.bat); input/output, config path, per-feature checkboxes
  • CLI – -i input, -o output filename
  • Obfuscate folder – Output written to Obfuscate/ in the obfuscator JAR directory
  • Overwrite – Replaces existing output file if present

Developer Experience

  • Colored logging – INFO, STEP, OK levels with colors (disable via NO_COLOR=1)
  • Version display – Shows version in startup banner
  • Gradle integration – Example project with obfuscate task
  • Test script – test-obfuscate.bat for quick end-to-end testing

Planned Features

Name Obfuscation

  • Overload induction – Multiple methods with same name, different signatures

Control Flow Obfuscation

  • Opaque predicates – Conditions that always evaluate true/false but are hard to analyze
  • Bogus control flow – Fake branches, dead code, unreachable blocks
  • Control flow flattening – Replace linear flow with switch/dispatch structure
  • Indirection – Wrap logic in extra method calls or trampolines
  • Loop transformation – Modify loop structure to obscure intent

String Obfuscation

  • String encryption ✓ (XOR-based, inline decrypt at each use site; no central decoder)
  • String splitting – Split strings across multiple concatenations
  • Encoding – Base64, XOR, custom encodings
  • Reflection hiding – Obfuscate strings used in reflection

Data Obfuscation

  • (none remaining)

Structural Obfuscation

  • Class splitting – Split large classes into smaller ones
  • Class merging – Merge unrelated classes
  • Fake classes – Add dummy classes/methods to mislead analysis
  • Annotation removal – Strip or obfuscate annotations (configurable)

Anti-Tampering & Anti-Debugging

  • Integrity checks – Detect modification via checksums/hashes
  • Anti-debugging – Detect debugger attachment, timing checks
  • Anti-dump – Hinder memory dumps or runtime inspection

Compatibility & Usability

  • Mapping file – Output deobfuscation mapping for stack traces
  • Keep rules – ProGuard-style keep rules for classes/methods/fields
  • Gradle/Maven plugins – Easy integration into build pipelines
  • Incremental obfuscation – Support incremental builds where possible