June 27, 2026
My AI Workflow Learned to Check Its Toolbelt
I added a large codebase readiness layer to awesome-ai-workflow so agents can decide when Serena-like tooling is worth it, check what is installed, and ask before touching your config.
Blog
Writing.
Technical articles and research write-ups. New posts publish here first, with older external writing linked below.
May 31, 2026
Your AI Workflow Needs a Flight Recorder
A practical agentic workflow that combines Superpowers, Memory Bank, Ralph-style loops, and dynamic workflow packets without turning every task into a paperwork simulator.
April 2, 2026
Ralph-NG: Your AI Shouldn't Grade Its Own Homework
Claude writes the code. Codex reviews it blind. If they disagree, they argue. If they're stuck, a human steps in. Here's how I built it.
March 28, 2026
The Hitchhiker's Guide to LLMs: From 'Hey ChatGPT' to Autonomous Coding Agents
Everything I know about using LLMs well. Not the hype version. Not the doom version. The practical one.
March 23, 2026
Coding Is Dead. Engineering Never Was.
The spec is the product now. Code is a commodity. Here's what that means and what I'm building about it.
March 18, 2026
Crossfire: When One LLM Isn't Enough
I built a web app that makes Claude and Codex argue with each other before writing a single line of code. The specs are better. The code is better. Here's how.
March 14, 2026
Your CAPTCHA Is Only as Strong as Your Weakest Model
How a math-based Django CAPTCHA fell apart the moment I pointed a vision LLM at it. Tesseract choked. Gemini didn't even blink.
March 11, 2026
Stop Committing Your Secrets (You Know Who You Are)
Plaintext .env files are a stupid little footgun. Here's the SOPS + age + direnv setup I use to keep secrets encrypted, auto-loaded, and out of Git.
March 1, 2026
The Interception Layer: A Terrible Idea That Might Actually Work
A middleware architecture for intercepting, anonymizing, and logging all data between autonomous AI agents and frontier LLMs.
February 28, 2026
Your Autonomous Pentest Agent Might Be the Biggest Vulnerability in Your Network
AI agents with root access, no NDA, no background check, and no clue where your data ends up. Why autonomous pentest platforms are a compliance and security nightmare.
July 15, 2025
Taming the Beast — Prompt Engineering and Agent Guardrails
Part 2 of the MITRE ATT&CK agent series — prompt engineering techniques, chain-of-evidence validation, and fixing infinite loops.
July 15, 2025
I Attempted to Build an Agentic AI... And It Immediately Got Stuck in a Loop
Part 1 of building a MITRE ATT&CK mapping agent crew — architecture, RAG pipeline, and the first spectacular failures.
June 23, 2025
Bring Your Own Fix — Mr.D0x inspired variation of yet another "fix" attack
A downloadfix PoC using Service Workers to simulate failed downloads and trick users into running a "repair tool."
April 30, 2025
Sorry if it's hard to catch my vibe — Building the Dumbest (Yet Smartest?) C2 in Existence
How well can GenAI deal with new trends? Building a minimalist C2 with dynamic LLM-assisted capability generation via MCP.
August 25, 2021
%appdata% is a mistake — Introducing Invoke-DLLClone
Combining DLL metadata cloning, export table copying, and fake code-signing into a single PowerShell workflow for DLL hijacking operations.
August 5, 2021
A pinch of XLL and a splash of Rust has the potential to be a sharp combination
Emulating the Buer Loader threat using Rust for reconnaissance and a C# XLL dropper — from zero Rust experience to a working PoC.
June 16, 2021
Click your shortcut and... you got pwned
Introducing LnkGen, a GUI tool for crafting malicious LNK shortcut files with bamboozle mode, alternate data streams, and expert options.
December 21, 2020
A tale of .NET assemblies, cobalt strike size constraints, and reflection
Exploring how reflection and AppDomain.AssemblyResolve can bypass Cobalt Strike's 1MB execute-assembly limit by loading .NET dependencies at runtime.
Archive
External writing
Published on NVISO Labs and other platforms.
SANS Blog
A Decade of Security Assessments: Security Issues That Refuse to Die
Recurring security issues observed over a decade of assessments — the problems that keep coming back.
Neuvik
Using GenAI to Encode Malware and Bypass EDR
Exploring how generative AI can be used to encode malware payloads and evade endpoint detection.
Neuvik
The Tale of Privilege Escalation vs the Unstartable Service
Technical case study on a privilege escalation finding involving a service that wouldn't start.
TrustedSec
A Comprehensive Guide on Relaying Anno 2022
In-depth guide covering modern NTLM relay attack techniques and mitigations.
NVISO Labs
Tap Tap, Is This Thing On? Creating a Notification Service for Cobalt Strike
Building a notification service to alert operators when new beacons check in on Cobalt Strike.
NVISO Labs
I Solemnly Swear I Am Up to No Good: Introducing the Marauders Map
Introducing a tool for mapping out attack paths and visualizing offensive operations.
SCYTHE
Using SCYTHE Payload as Shellcode
Walkthrough on converting SCYTHE payloads into shellcode for use in custom loaders.
NVISO Labs
The Return of the Spoof Part 1: Parent Process ID Spoofing
Techniques for spoofing parent process IDs to evade detection during offensive operations.
NVISO Labs
Unmanaged File Searching with FileSearcher.exe
Using an unmanaged .NET file searcher to locate sensitive files during red team engagements.