Technical write-ups on AI security, LLM pentesting, and supply-chain vulnerability research.https://cooties.io/https://cooties.io/blog/anatomy/https://cooties.io/blog/anatomy/A meta-analysis of jailbreak tactics, why a frame has to fit the request, and the role of context in multi-turn attacks.Fri, 02 Oct 2026 00:00:00 GMTAI safetyjailbreaksllmai-safetyredteaminterpretabilityhttps://cooties.io/blog/project-umbra/https://cooties.io/blog/project-umbra/Building an agentic vulnerability scanner, finding a real backdoor, and why this kind of work should be standard practice.Tue, 21 Apr 2026 00:00:00 GMTCybersecuritysupply-chainllmvulnerability-researchpentestinghttps://cooties.io/blog/mr-bones-wild-ride/https://cooties.io/blog/mr-bones-wild-ride/Agentic AI, labor economics, and the end of the apprenticeship model.Wed, 25 Mar 2026 00:00:00 GMTAI safetyopinioncybersecurityeconomicsai-safetyhttps://cooties.io/blog/pentesting-llm-applications-part2/https://cooties.io/blog/pentesting-llm-applications-part2/Deep dive into prompt injection attacks, defense bypasses, and exploiting LLM applications - from basic injections to advanced multi-stage attacks combining AI-specific and traditional vulnerabilities.Sun, 14 Dec 2025 00:00:00 GMTCybersecuritypentestingllmsecurityprompt-injectionredteamexploitationhttps://cooties.io/blog/pentesting-llm-applications/https://cooties.io/blog/pentesting-llm-applications/Understanding the unique architectural properties that make LLMs challenging to secure, distinguishing AI safety from security, and mapping the attack surface through reconnaissance techniques.Sun, 07 Dec 2025 00:00:00 GMTCybersecuritypentestingllmsecurityprompt-injection