LLM & Agent Exploitation
Deep dives into prompt injection primitives, RAG poisoning, MCP tool hijacking, memory persistence, and autonomous agent takeovers.
Guardrail & Evasion Engineering
Analysis of alignment bypasses, classifier probing, SmoothLLM mechanics, and protocol-layer vulnerabilities in production AI.
Red Team Tradecraft & Tools
Automated attack frameworks (GPTFuzzer, PAIR, J2 paradigm), C2 architectures, zero-day analysis, and working PoCs.
1 issue published
🔒

On privacy: Your email is stored solely for delivering this newsletter. It is never sold, shared with third parties, or used for tracking. Unsubscribe at any time with a single click. There are no corporate sponsors or advertiser relationships, and security knowledge should move freely.

See what I'm building right now →