Long-form notes on security, protocol research, AI-driven engineering, and the machines in between.https://michael.mccord.ai/https://michael.mccord.ai/blog/plaintext-cut/https://michael.mccord.ai/blog/plaintext-cut/A design weakness in where semantic monitors sit in agent pipelines, written up as an advisory: the necessary condition a control has to satisfy, a working proof of concept, an interactive coverage model, and a procedure for testing your own deployment. The remediation is not a better classifier.Thu, 06 Aug 2026 00:00:00 GMTAI security research · advisoryhttps://michael.mccord.ai/blog/lamiai/https://michael.mccord.ai/blog/lamiai/AI attacks are scaling past what patching and detection can cover. But an AI agent's eagerness to finish its task is a weakness you can use: drop it into a decoy environment and read what it is after from what it does, no prompt injection to fail or tip it off. Real captures, cited threat reporting, and open code meant to go upstream into Beelzebub.Wed, 01 Jul 2026 00:00:00 GMTSecurity research · open sourcehttps://michael.mccord.ai/blog/fingerprinting-abyss/https://michael.mccord.ai/blog/fingerprinting-abyss/Cloudflare's new PACT proposal says inspecting browsers to catch bots has failed. I reverse-engineered a vendor that bets the other way, to test how much weight fingerprinting can still carry.Mon, 01 Jun 2026 00:00:00 GMTSecurity researchhttps://michael.mccord.ai/blog/lotus-exige-track-build/https://michael.mccord.ai/blog/lotus-exige-track-build/Four years turning a used, near-stock Lotus Exige S into a track car — and turning myself into a driver who understood why every change was a trade-off.Fri, 01 May 2026 00:00:00 GMTThe Garage Files № 01