How Frida Works?

Purchase now

Stop just using dynamic instrumentation tools. Learn how to build them from scratch.

We are incredibly excited to announce the presale for our newest Fuzzsociety deep-dive: How Frida and Debuggers Work: Building a Lua-Based Injector, led by Antonio Nappa.

In this ~10-hour intensive course, we are pulling back the curtain on how tools like Frida actually function under the hood. You won’t just study the theory—you will build a complete, working Lua-based injector from the ground up using C.

No credits available

Here is exactly what you will build and learn:

Phase 1: Debugger Internals and Primitives

Master ptrace and OS-level debugging APIs.
Implement software/hardware breakpoints and single-stepping.
Lab: Write a minimal debugger in C from scratch.

Phase 2: Code Injection in C

Explore LD_PRELOAD, manual mapping, and trampoline hooks.
Master GOT/PLT hijacking directly at the C level.
Lab: Write a function hook without relying on any external frameworks.

Phase 3: Deconstructing Frida

Reverse engineer Frida’s core architecture: GumJS, QuickJS, Stalker, and Interceptor.
Map the agent/host channel model to see exactly how Frida injects and communicates on live targets.

Phase 4 and 5: Building Your Custom Scripting Engine

Learn Lua embedding fundamentals and manage the C API/stack model.
Expose your native C instrumentation primitives (hooks, memory, registers) to Lua.
Lab: Build the complete injector architecture (loader, C runtime core, Lua script host) and write instrumentation scripts against a real target.

Who is this for?

Vulnerability researchers and engineers with intermediate C programming, familiarity with process memory layouts, and x86/x64 assembly experience. (Lab environment runs on Ubuntu 22.04 LTS).