Skip to content
All projects
01Actively developed

Personal Project · Rust · Security Tooling · AI

Farstyle — Security Auditing Platform

A native Rust desktop auditing platform: interception proxy, repeater, intruder, a pluggable module engine and an AI workspace, in a single binary.

Rustegui / eframerustlsPythonAI

Overview

Farstyle is a native Rust desktop security auditing platform built with egui/eframe. It bundles an HTTP/HTTPS interception proxy, a repeater, an intruder, a module engine, an AI workspace and pluggable security scanners into a single application.

Objective

Build a working, self-contained auditing tool from the ground up — to understand how professional-grade tooling (proxies, TLS interception, fuzzing engines) actually works internally, rather than only using it as a black box.

Architecture

Module pipeline

How a request flows from the operator down to the underlying tools.

User

Operator sets a goal or crafts a request in the GUI.

Interception path

The HTTP/HTTPS traffic path through the interception proxy.

HTTP Client

The browser or client whose traffic is under test.

Technologies

  • Rust
  • egui / eframe
  • rustls
  • rcgen
  • Python modules
  • JSON-over-stdin/stdout IPC
  • BM25 retrieval

What I built

  • A native Rust desktop app (egui/eframe) bundling a proxy, repeater, intruder, module engine and AI workspace in a single binary.
  • An HTTP/HTTPS intercepting proxy using rustls and rcgen for on-the-fly TLS certificate generation.
  • A Python module system with a JSON-over-stdin/stdout IPC protocol for pluggable scanners.
  • Built-in modules for reconnaissance, fuzzing and SQL injection.
  • An AI workspace that maps plain-English goals to the right tool or module (nmap, ffuf, sqlmap, nuclei, subfinder, custom modules) using BM25 retrieval over a knowledge base.

Security concepts

  • TLS interception & certificate generation
  • Proxy architecture and request/response tampering
  • Fuzzing and injection payload engines
  • Inter-process communication for plugin isolation
  • Retrieval-driven tool selection

What I learned

  • How an interception proxy establishes trust and rewrites TLS traffic in the middle.
  • Designing a plugin boundary (IPC) so modules can be written and iterated in Python without touching the Rust core.
  • Mapping fuzzy, natural-language intent onto a concrete, bounded set of security tools with retrieval.

Actively developed — not a production security tool, and not commercially deployed.

Next project

Vulnerable Web Application Lab