Rob Ragan

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

offensive securityAI agentspenetration testingred teamingapplication security

1135 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. At Starlog, he curates expert deep-dives on the open-source tools reshaping offensive security and AI agent development.

// ARTICLES BY ROB RAGAN

Developer Tools

SRMT: Teaching Robot Swarms to Navigate Using Shared Attention Instead of Radios

By Rob Ragan ★ 34 Python Apr 3, 2026

Developer Tools

AIVSS: Adapting CVSS Vulnerability Scoring for the Age of AI Systems

By Rob Ragan ★ 38 Python Apr 3, 2026

Developer Tools

LLM-Check: Detecting Hallucinations by Analyzing What Language Models Think, Not Just What They Say

By Rob Ragan ★ 39 Jupyter Notebook Apr 3, 2026

Developer Tools

Web-Shepherd: Training Process Reward Models to Guide Web Agents Through Long-Horizon Tasks

By Rob Ragan ★ 53 Python Apr 3, 2026

Developer Tools

PenGym: Training Reinforcement Learning Agents on Real Penetration Testing Infrastructure

By Rob Ragan ★ 56 Python Apr 3, 2026

Developer Tools

SEC-bench: Automated Benchmarking for LLM Agents on Real-World Security Vulnerabilities

By Rob Ragan ★ 63 Python Apr 3, 2026

Developer Tools

HackBench: The Security Benchmark Where LLMs Learn to Exploit Real Vulnerabilities

By Rob Ragan ★ 69 Rich Text Format Apr 3, 2026

Developer Tools

InterCode-CTF: How Simple Prompts Cracked 95% of Security Challenges (And What That Means for LLM Benchmarking)

By Rob Ragan ★ 70 Python Apr 3, 2026

Rob Ragan

// ARTICLES BY ROB RAGAN

SRMT: Teaching Robot Swarms to Navigate Using Shared Attention Instead of Radios

AIVSS: Adapting CVSS Vulnerability Scoring for the Age of AI Systems

LLM-Check: Detecting Hallucinations by Analyzing What Language Models Think, Not Just What They Say

AgentSonar: Detecting Shadow AI Through Network Traffic Heuristics

Building RAG Chat with Exa's Web Search API and OpenAI o3-mini

Vibe Security Radar: Tracking CVEs Caused by AI-Generated Code

Shift: Teaching AI to Manipulate HTTP Traffic Like a Penetration Tester

Web-Shepherd: Training Process Reward Models to Guide Web Agents Through Long-Horizon Tasks

PenGym: Training Reinforcement Learning Agents on Real Penetration Testing Infrastructure

SEC-bench: Automated Benchmarking for LLM Agents on Real-World Security Vulnerabilities

HackBench: The Security Benchmark Where LLMs Learn to Exploit Real Vulnerabilities

InterCode-CTF: How Simple Prompts Cracked 95% of Security Challenges (And What That Means for LLM Benchmarking)