<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>mvonal.dev</title><description>Engineering lab notebook. Portfolio, writing and a knowledge graph in one place.</description><link>https://mvonal.dev/</link><language>en-us</language><item><title>AI Security Homelab Part 1: I Built a Vulnerable AI Infrastructure and Pentested It (Reconnaissance)</title><link>https://mvonal.dev/articles/ai-security-homelab-part-1/</link><guid isPermaLink="true">https://mvonal.dev/articles/ai-security-homelab-part-1/</guid><description>First post in a series on building an isolated AI security homelab: how the lab was set up, then Phase 1, reconnaissance against a deliberately vulnerable AI infrastructure stack.</description><pubDate>Tue, 11 Aug 2026 00:00:00 GMT</pubDate></item><item><title>AI Security Homelab Part 2: I Built a RAG Chatbot, Then Leaked Everything Inside It</title><link>https://mvonal.dev/articles/ai-security-homelab-part-2/</link><guid isPermaLink="true">https://mvonal.dev/articles/ai-security-homelab-part-2/</guid><description>I built a real RAG chatbot on my vulnerable AI homelab, then reran recon on the live system: the same secrets leaked from Qdrant, MLflow, and MinIO, plus a planted prompt injection seed.</description><pubDate>Thu, 13 Aug 2026 00:00:00 GMT</pubDate></item><item><title>AI Security Homelab Part 3: Attacking the RAG Chatbot (OWASP LLM Top 10)</title><link>https://mvonal.dev/articles/ai-security-homelab-part-3/</link><guid isPermaLink="true">https://mvonal.dev/articles/ai-security-homelab-part-3/</guid><description>Third post in the AI Security Homelab series: attacking my own vulnerable RAG chatbot through the OWASP LLM Top 10, leaking its prompt, chaining to stored XSS, and automating with garak.</description><pubDate>Fri, 14 Aug 2026 00:00:00 GMT</pubDate></item><item><title>AI Security Homelab Part 4: Building the Defenses</title><link>https://mvonal.dev/articles/ai-security-homelab-part-4/</link><guid isPermaLink="true">https://mvonal.dev/articles/ai-security-homelab-part-4/</guid><description>The character of this phase is before and after. For every finding from Parts 1 through 3, I put a control in front of it, then rerun the exact attack and prove it now fails, whether that means 401…</description><pubDate>Fri, 21 Aug 2026 00:00:00 GMT</pubDate></item><item><title>AI Security Homelab Part 5: Attacking My Own Defenses</title><link>https://mvonal.dev/articles/ai-security-homelab-part-5/</link><guid isPermaLink="true">https://mvonal.dev/articles/ai-security-homelab-part-5/</guid><description>Up to now we tested each defense by asking “does a harmless request still get through.” This phase reverses that. We try to defeat each defense deliberately, and we record what it stopped and what…</description><pubDate>Fri, 21 Aug 2026 00:00:00 GMT</pubDate></item><item><title>AI Security Homelab Part 6: Building a SIEM and Writing Detections</title><link>https://mvonal.dev/articles/ai-security-homelab-part-6/</link><guid isPermaLink="true">https://mvonal.dev/articles/ai-security-homelab-part-6/</guid><description>The goal of this phase was not to become a SOC analyst. It was to understand, at a practical level, how a SIEM actually works: how raw logs flow in, what they look like before anyone makes sense of…</description><pubDate>Mon, 24 Aug 2026 00:00:00 GMT</pubDate></item><item><title>AI Security Homelab Part 7: Attacking an AI Agent</title><link>https://mvonal.dev/articles/ai-security-homelab-part-7/</link><guid isPermaLink="true">https://mvonal.dev/articles/ai-security-homelab-part-7/</guid><description>Everything runs in an internet isolated virtual network on my own machine.</description><pubDate>Mon, 31 Aug 2026 00:00:00 GMT</pubDate></item><item><title>AI Security Homelab</title><link>https://mvonal.dev/projects/ai-security-homelab/</link><guid isPermaLink="true">https://mvonal.dev/projects/ai-security-homelab/</guid><description>An isolated homelab for attacking and hardening AI/ML infrastructure: attack, harden, attack again, document.</description><pubDate>Sat, 01 Aug 2026 00:00:00 GMT</pubDate></item><item><title>UAV Telemetry Fuzzer</title><link>https://mvonal.dev/projects/mavlink-fuzzer/</link><guid isPermaLink="true">https://mvonal.dev/projects/mavlink-fuzzer/</guid><description>A MAVLink protocol fuzzer written in C that targets ArduPilot SITL, with bit-flip, boundary-value and length-corruption strategies.</description><pubDate>Sat, 04 Jul 2026 00:00:00 GMT</pubDate></item><item><title>Obsidian Local Brain</title><link>https://mvonal.dev/projects/obsidian-local-brain/</link><guid isPermaLink="true">https://mvonal.dev/projects/obsidian-local-brain/</guid><description>A fully local RAG pipeline that turns an Obsidian vault into daily spaced-repetition quizzes with a local vision LLM.</description><pubDate>Sat, 01 Nov 2025 00:00:00 GMT</pubDate></item><item><title>SMS Spam Detection</title><link>https://mvonal.dev/projects/sms-spam-detection/</link><guid isPermaLink="true">https://mvonal.dev/projects/sms-spam-detection/</guid><description>An end-to-end SMS spam classifier built with classical NLP and Multinomial Naive Bayes.</description><pubDate>Thu, 01 Jan 2026 00:00:00 GMT</pubDate></item><item><title>WebSift: Context Engineering Lab</title><link>https://mvonal.dev/projects/websift/</link><guid isPermaLink="true">https://mvonal.dev/projects/websift/</guid><description>One research agent, three architectures, one honest benchmark: how much context choices cost in tokens, calls, latency and faithfulness.</description><pubDate>Tue, 01 Sep 2026 00:00:00 GMT</pubDate></item><item><title>YamanSec</title><link>https://mvonal.dev/projects/yamansec/</link><guid isPermaLink="true">https://mvonal.dev/projects/yamansec/</guid><description>A modular web vulnerability scanner with a Flask UI, a custom SQLi/XSS engine, OWASP ZAP integration and PDF reports.</description><pubDate>Sun, 01 Jun 2025 00:00:00 GMT</pubDate></item></channel></rss>