Rob Ragan

Offensive security researcher and AI agent architect. Founder of Starlog.

offensive securityAI agentspenetration testingred teamingapplication security

1756 articles

Rob Ragan is an offensive security researcher and AI agent architect who has spent two decades breaking into systems and building tools to automate the process.

// ARTICLES BY ROB RAGAN

Developer Tools

Natural Language Autoencoders: Reading Neural Activation Vectors as English Text

★ 895 Python Jul 13, 2026

AI Agents

Building an MCP Server for Apple's TestFlight Feedback: Why JWT Caching Matters

★ 1 Python Jul 12, 2026

LLM Engineering

Running 753B MoE Models on Consumer GPUs: Hand-Written SASS Kernels for 2-Bit Experts

★ 253 Sass Jul 10, 2026

AI Agents

Agentic Awesome Skills: The npm Registry That AI Coding Agents Actually Needed

★ 42.6k Python Jul 8, 2026

Developer Tools

Watching Language Models Think: Real-Time Jacobian Lens Visualization with Subtext

★ 32 HTML Jul 7, 2026

AI Agents

Fable5: Engineering AI Agent Reliability Through Enforcement, Not Hope

★ 18 Shell Jul 7, 2026

Cybersecurity

Exploitarium: A 0-Day Archive That Weaponizes AI-Assisted Fuzzing Against 35+ Production Systems

★ 3.7k Python Jul 7, 2026

AI Agents

Shepherd: Git-Like Forking for Agent Execution Traces

★ 626 Python Jul 6, 2026

AI Dev Tools

Engram: The AI Tutor That Finally Solves Self-Grading With Architectural Isolation

★ 216 Python Jul 6, 2026

AI Agents

Rob Ragan

// ARTICLES BY ROB RAGAN

Natural Language Autoencoders: Reading Neural Activation Vectors as English Text

Building an MCP Server for Apple's TestFlight Feedback: Why JWT Caching Matters

Running 753B MoE Models on Consumer GPUs: Hand-Written SASS Kernels for 2-Bit Experts

Agentic Awesome Skills: The npm Registry That AI Coding Agents Actually Needed

Watching Language Models Think: Real-Time Jacobian Lens Visualization with Subtext

Fable5: Engineering AI Agent Reliability Through Enforcement, Not Hope

Exploitarium: A 0-Day Archive That Weaponizes AI-Assisted Fuzzing Against 35+ Production Systems

Shepherd: Git-Like Forking for Agent Execution Traces

Engram: The AI Tutor That Finally Solves Self-Grading With Architectural Isolation

PYTHIA: Building a Self-Calibrating Forecast Oracle with Local LLMs

FluxCapacitor: A Git-Aware Filesystem Timeline for Supervising AI Coding Agents

T3MP3ST: The Offensive Security Framework That Actually Shows Its Work