Reverse Engineering logo

Reverse Engineering

CommunityPopular
rehan-remade
reverse-engineering

Read how a game actually works so a mod can hook it. Decompile .NET/Mono (ILSpy), IL2CPP (Cpp2IL, Il2CppDumper), Java (Vineflower) and native code (Ghidra, IDA, via MCP servers); dump data and asset formats; reverse-engineer a binary file format and prove it with a round trip; read live memory (Cheat Engine, Frida, x64dbg); find render passes (RenderDoc). Use when a mod needs game internals, e.g. "how does the boss AI work", "find the damage function", "what format is this .sld/.pak/.dat", "where is the player position in memory".

Overview

Publisherrehan-remade
Repositoryuniversal-modder
Skill namereverse-engineering
Stars
3.4K
Forks
293
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by rehan-remade on GitHub. Read the source before you install it.

Installation

Install the Reverse Engineering AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/rehan-remade/universal-modder/tree/main/.agents/skills/reverse-engineering
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/rehan-remade/universal-modder.git /tmp/universal-modder
mkdir -p .claude/skills
cp -r /tmp/universal-modder/.agents/skills/reverse-engineering .claude/skills/reverse-engineering
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Reverse Engineering in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Reverse Engineering on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Reverse Engineering is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Reverse engineering for mods

Rule zero: read the real thing, don't guess. Decompiled code, the actual data, a live memory view or a GPU capture is the spec. Write what you learn into MODLOG.md (names, IDs, offsets, formats) as you go.

Keep all decompiled output and extracted assets outside the mod repo (e.g. ~/<game>-decomp/) and never publish them. Single-player, offline or servers the user runs. Never attach debuggers or scanners to games with anti-cheat (see skills/mod-any-game/references/safety.md).

Pick the tool by what the code is (um scan tells you)

Managed .NET (XNA/FNA, Unity Mono, most indie C#)

  • dotnet tool install -g ilspycmd, then ilspycmd -p -o ~/<game>-decomp <Game>.exe (or Assembly-CSharp.dll). This gives a full C# project you can grep.
  • dnSpyEx: step through with a debugger, set breakpoints, edit methods.
  • MCP: ILSpy-MCP / dnspy-mcp let an agent query types and methods directly.

Unity IL2CPP

  • Cpp2IL (or Il2CppDumper) on GameAssembly.dll + global-metadata.dat gives types, fields, method signatures and addresses, plus dummy DLLs for ILSpy.
  • Method bodies are native: load GameAssembly.dll in Ghidra/IDA and apply the generated script to name functions.
  • Live: UnityExplorer, or il2cpp-frida-mcp.

Java (Minecraft, Slay the Spire...) Vineflower / CFR / Recaf. For Minecraft, use Loom genSources with Mojang mappings.

Native C/C++ (custom engines, Unreal game code, console recomps)

  • Ghidra (free) or IDA, driven through MCP so the agent can decompile, rename, retype and follow cross-references:
    • Ghidra: GhidraMCP (LaurieWired), pyghidra-mcp (headless, with a run-script tool; code-capable tools beat hundreds of tiny ones), ReVa.
    • IDA: the official Hex-Rays IDA MCP (IDA 9.4+ Pro/Home; "code mode" runs IDAPython), or ida-pro-mcp.
    • Binary Ninja and radare2 have MCP servers too.
  • Workflow:
    1. Strings, then cross-references, then the function.
    2. Name and type everything you understand. Renames accumulate into a readable program.
    3. Confirm dynamically (below) before building on a guess. Agents can confidently misidentify things.
  • Unreal: dump the reflection data first (UE4SS dumper or Dumper-7); it names most gameplay classes and properties for free.

Dynamic / live

  • Cheat Engine: value scans → "find out what writes to this address" → struct → owner. CheatEngine MCP servers exist.
  • x64dbg: breakpoints and tracing (x64dbg-mcp; bind it to 127.0.0.1, since some default to 0.0.0.0).
  • Frida (frida-mcp, frida-game-hacking-mcp): hook functions from JavaScript, log arguments.
  • ReClass.NET rebuilds structs from live memory.

Graphics

  • RenderDoc (renderdoc-mcp) captures a frame and shows every draw, the render targets, the constant buffers (view/projection matrices) and the depth buffer.
  • That's how you find where to inject geometry or effects (mashup-mods), and which texture holds a sprite atlas.

Data files and asset formats

Use the community tool first:

  • Unity: UABEA, AssetRipper
  • Unreal: FModel, UAssetGUI
  • Bethesda: xEdit, BSArch
  • GameMaker: UndertaleModTool
  • Genie: genieutils
  • Source: VRF, Crowbar
  • FromSoft: WitchyBND, Smithbox
  • Godot: GDRE Tools

Before reversing a format yourself, search the community archives. Most game formats from the 2000s and 2010s were worked out there:

Use them for format knowledge and tools, whatever its origin (betas, leaked SDKs). Don't download game files, leaked code, SDKs or license keys from them.

For an undocumented format:

  1. Collect several stock files. Compare sizes, and hex-dump the headers (xxd | head). Look for magic numbers, counts, offsets and tables of fixed-size records.
  2. Form a hypothesis for the header, then the frame/record layout, then the compression. Check it by parsing every stock file without errors.
  3. Write a reader that decodes to something viewable (PNGs, JSON) and look at it.
  4. Write the writer, and prove it with a round trip: decode → encode → decode, compared against the original. The AoE2 SLD sprite writer was accepted only when it round-tripped the stock knight at 0.9/255 mean error (examples/aoe2-de-civ/sld.py).
  5. Only then write new files. Test them in game with one asset before batch-converting.

Make it an oracle

  • Engine logic you port (for a simulator, a trainer, a reimplementation): record real traces from the game (positions, velocities per tick) and replay them against your port. In the Terraria Eye of Cthulhu work, the Eye's velocity matched 99.9% once the port used the action applied on frame t+1. Float32 constants mattered too (0.2f ≠ 0.2). A leftover mismatch was traced to a hidden buff (Happy!, x1.21 move speed). Replays find what reading the code misses.
  • For long RE runs:
    • a journal file;
    • small verified steps;
    • cap attempts per problem (about 3 identical failures, then change approach);
    • commit every confirmed fact.

Frequently asked questions

What does the Reverse Engineering AI skill do?

Read how a game actually works so a mod can hook it. Decompile .NET/Mono (ILSpy), IL2CPP (Cpp2IL, Il2CppDumper), Java (Vineflower) and native code (Ghidra, IDA, via MCP servers); dump data and asset formats; reverse-engineer a binary file format and prove it with a round trip; read live memory (Cheat Engine, Frida, x64dbg); find render passes (RenderDoc). Use when a mod needs game internals, e.g. "how does the boss AI work", "find the damage function", "what format is this .sld/.pak/.dat", "where is the player position in memory".

Why use Reverse Engineering on TypingMind?

Because you install it once and use it with any model. Reverse Engineering is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Reverse Engineering in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/rehan-remade/universal-modder/tree/main/.agents/skills/reverse-engineering. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Reverse Engineering?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Reverse Engineering?

As many as you like. As long as a model supports skills, you can use Reverse Engineering with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Reverse Engineering AI skill free?

Yes. It is published on GitHub by rehan-remade under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇