Inspect and record HTTP traffic with an interactive proxy, web interface, or scriptable command-line tools.
Tool · Foundation tool · App, Website
Requirements and limits
Configure the client proxy; HTTPS inspection requires the appropriate local trust configuration. Saved captures and live interception are distinct workflows.
Recover readable modules from shipped JavaScript bundles by reversing recognized minifier and transpiler patterns. Supports bundle splitting and modern-syntax recovery for inspecting production frontend code.
Tool · Foundation tool · Website
Requirements and limits
The npm CLI declares Node.js >=16; standalone v1.14.0 binaries are published for macOS arm64, Linux arm64/x64, and Windows arm64/x64.
Provide supported JavaScript bundles or source maps. Original identifiers may be unrecoverable without maps; heavy control-flow or VM obfuscation requires other tooling. Package/version inventory is explicitly still in development.
Analyze captured application HTTP traffic with an LLM to generate MCP tools, OpenAPI 3.1 specifications, or GraphQL schemas. Generated tools call the observed API directly after capture and authentication setup.
Python >=3.11; capture through the Chrome extension or MITM proxy, and configure a supported LLM provider or local Ollama endpoint.
The reviewed v0.11.0 workflow depends on captured traffic and valid application authentication. Analysis and automatic repair use the configured LLM; this entry does not claim universal API coverage or verified generated-tool correctness.
Inspect JavaScript ASTs and variable scopes while building visual chains of Babel transforms. Save analysis projects and reusable plugins to iteratively simplify obfuscated scripts.
Tool · Foundation tool · Website
Requirements and limits
Run from source with Node.js and npm; the documented default standalone build targets Windows. No GitHub release was available at review time.
Built-in transforms may need sample-specific scripts, and deeply nested input can overflow the stack. Its optional evaluation helper executes code and should be used in a local sandbox, as upstream explicitly advises.
Browse and modify supported GameMaker data files, extract assets, and decompile or compile GML VM code. Includes room editing, scripted changes, a CLI, and the reusable UndertaleModLib library.
Tool · Foundation tool · Game
Requirements and limits
Supply a supported game data file such as data.win, game.ios, or game.unx. Stable 0.9.2.0 provides a 64-bit Windows GUI and CLI packages for Windows, Ubuntu, and macOS.
GML VM code editing excludes YYC, and some language features remain unsupported. Building the reviewed source requires the .NET 10 SDK or later and recursive submodules; the GUI build requires Windows.
Inspect IL2CPP classes, methods, and fields in a live process, and trace or intercept method calls without supplying global-metadata.dat. Complements static IL2CPP dumpers with runtime observation.
Tool · Foundation tool · App, Game
Requirements and limits
Requires Frida access to a running IL2CPP application and the npm module. The current bundled CLI needs Python 3.12+ (inferred from its typing.override import) and Frida tooling.
The reviewed README states Unity 5.3.0–6000.3.x support. Android and Linux are the platforms upstream tests; Windows, iOS, and macOS are listed as supported with explicit warnings of possible breakage. Latest release at review: v0.14.0.
Compare functions and data in reference and rebuilt object files to guide matching decompilation. Displays instruction-level differences, demangles C++ symbols, and can rebuild when source files change.
Tool · Foundation tool · Binary, Game
Requirements and limits
Provide reference and rebuilt object files. Project integration and automatic rebuilding use objdiff.json plus a working toolchain and build command; direct CLI comparisons can use file paths.
v3.8.2 provides GUI binaries for Windows/Linux x86_64 and macOS arm64/x86_64, with additional CLI targets. Building from source requires Rust. Supported architectures include ARM, ARM64, MIPS, PowerPC, SuperH, and x86/x86_64.
Explore Unreal packages through class, property, and dependency trees; decompile and export UnrealScript classes, inspect bytecode, and examine structured binary data.
Tool · Foundation tool · Binary, Game
Requirements and limits
Windows application built on .NET Framework 4.8; provide supported Unreal Engine 1, 2/2.5, or 3 packages. Latest stable release at review: 1.6.2.
UE2X and UE4/5 are not generally supported; only limited experimental parsing of loose UE4 .uasset containers is documented. Game-specific engine modifications can affect compatibility; older 1.6-and-earlier builds require decompressed packages.
Browse recognized visual-novel resource archives as folders, preview their contents, and extract images or audio with conversion to common formats.
Tool · Foundation tool · Game
Requirements and limits
Requires Windows and .NET Framework 4.6 or newer; archive browsing and conversion depend on the supported game/format list. Encrypted archives may require credentials or a recognized game selection.
Latest GitHub release at review remains v1.5.44 (2019-03-31). Treat it as a legacy format tool and check format support for the specific title rather than assuming compatibility with newer games.
Inspect mobile application packages with static analysis and supported-device dynamic testing, including runtime behavior and network traffic, through a web interface and APIs.
Tool · Foundation tool · App
Requirements and limits
The documented container setup recommends Docker 20.10+; dynamic analysis additionally needs a supported device or emulator accessible from the container.
Android dynamic analysis is documented for rooted Android 4.1–11, up to API 30; Android 12+ is not supported by that workflow.
iOS dynamic analysis uses Corellium or a jailbroken device; the physical-device route is marked beta.
Explore running Android and iOS apps with a Frida-backed command interface for inspecting app storage, memory, runtime objects, and instrumented behavior.
Tool · Foundation tool · App
Requirements and limits
Current package metadata requires Python 3.11+ and Frida 16+; this supersedes the older Python minimum still shown in the installation wiki.
The target must support the documented Frida attachment or patched-app workflow. Android patching requires Android build tools and Apktool 2.9.0+; iOS patching requires the documented Xcode and signing tools.
Use radare2 commands and Frida instrumentation together to inspect local or remote process memory, symbols, classes, and runtime calls on desktop and mobile targets.
Tool · Foundation tool · App, Binary
Requirements and limits
Install with r2pm or use a plugin build matching the radare2 major/minor ABI; the reviewed README reports CI against radare2 6.2.2 and git master.
Mobile or remote operation needs a reachable Frida target, such as frida-server over USB or TCP. The plugin bundles the host Frida library; separate host frida-tools are not required.
Extract firmware, combine static checks with emulation-based analysis, inventory software components, and generate web reports and SBOMs for further investigation.
Tool · Foundation tool · Firmware
Requirements and limits
Upstream recommends an x86-64 Kali Linux VM and the default Docker mode; Ubuntu 22.04/24.04 is also listed. Native ARM execution is not supported.
Documented resource guidance is at least 16 GB RAM, 8 CPU cores recommended, and 30–100 GB free disk space; larger firmware or more parallel workers need more resources.
Install the external dependencies through the project installer. Emulation can execute firmware content; upstream reserves developer mode for development, and findings still require interpretation.
Unpack and analyze firmware through an extensible platform whose web interface and HTTP API make component contents and analysis results searchable and comparable across samples.
Tool · Foundation tool · Firmware
Requirements and limits
The reviewed installer supports Debian 13 and Ubuntu 22.04/24.04/26.04; Kali support is experimental. The README lists Python 3.10–3.14.
Minimum documented resources are 4 CPU cores, 8 GB RAM, and about 10 GB for installation, plus storage for unpacked firmware and results.
The standalone FACT_docker distribution is marked unmaintained; use the documented local or Vagrant setup. Some analysis plugins have separate licenses.
Recursively identify, unpack, analyze, modify, and repack firmware and binary formats using a resource-tree GUI and Python API, with optional disassembler backends.
Tool · Foundation tool · Binary, Firmware
Requirements and limits
The PyPI installation guide requires Python 3.9+ and format-specific external tools; macOS also needs cmake and libmagic, avoiding the documented libmagic 5.47 regression.
Docker builds require BuildKit; Docker 23+ enables it by default. Analysis backends such as Ghidra or Binary Ninja have their own installation and licensing requirements.
OFRAK uses its own Community License for personal and educational use; commercial users should review the separate Pro or Enterprise licensing terms.
Emulate router and IP-camera firmware to inspect network services and runtime behavior, with automated emulation checks and separate analysis, run, and debugging modes.
Tool · Foundation tool · Firmware
Requirements and limits
The README documents Ubuntu 20.04 as the tested host; prepare the recursive source checkout, downloaded dependencies, initialized database, and a supported firmware image.
The installer configures QEMU, PostgreSQL, extraction tools, and other host packages; documented run modes require elevated privileges. Docker orchestration is also provided.
Successful emulation depends on the firmware image; upstream benchmark results are not a guarantee for an untested device or newer host distribution.
A bilingual collection of REA and other AI-assisted reverse engineering cases, with source links and explicit distinctions between reported and reproduced outcomes.
Collection · Community REA learning resource · App, Binary, Firmware, Game, Website
Requirements and limits
A discovery index; review each case's original source, redistribution rights, and evidence status before reproducing it.
Research on direct binary decompilation and refinement of Ghidra output with language models. Evaluates re-executability on HumanEval and ExeBench; useful for distinguishing readable output from functional recovery.
Research · Learning resource · Binary
Requirements and limits
EMNLP 2024 paper. Results are author-reported for the evaluated models and benchmarks; this collection has not reproduced them.
A benchmark combining functions from real programs, runtime-aware validation, and LLM-assisted assessment of decompiled-code usability. Helps separate functional correctness from perceived understandability when comparing decompilers.
Research · Learning resource · Binary
Requirements and limits
Findings of ACL 2025 paper. LLM-as-judge assessments and runtime correctness measure different properties; published results are not independent reproductions by this collection.
Combines language models and program analysis to recover variable and user-defined data-structure symbols from stripped binaries. The authors provide training, inference, evaluation, and posterior-reasoning artifacts.
Research · Learning resource · Binary
Requirements and limits
CCS 2024 paper with author-provided artifacts. Symbol and type recovery is a narrower task than reconstructing an entire original program; artifacts were not executed for this listing.