SLSean Leeinseanleecoder.hashnode.dev·1d ago · 11 min readIntroducing Rule-Trace: Your Agent Loaded the Rule. Did It Apply It?In my last post, I described a setup I had started using at work: give every AI-agent rule a stable ID, then ask the agent to disclose which rules were candidates and which ones actually shaped the re20
SLSean Leeinseanleecoder.hashnode.dev·6d ago · 14 min readMCP 2.0? Two Doorways to the Agentic UI — MCP Apps and WebMCPThe Model Context Protocol gave AI models a clean way to reach tools and data. What it didn't give them was an interface. In 2026 that changed twice over. MCP Apps lets an MCP server return rich, inte30
SLSean Leeinseanleecoder.hashnode.dev·May 12 · 8 min readSharing rules and slash commands across AI coding agentsAgent rules are easy to put in the wrong place. If you use one AI coding agent long enough, you accumulate a private operating manual for it: house style, review preferences, testing rules, package bo20
SLSean Leeinseanleecoder.hashnode.dev·May 12 · 9 min readMaking AI agent rule application visible — stable IDs and trace blocksYou wrote the rule. You loaded it into your AI agent. Did the agent use it? You can't tell. A rule that was loaded and ignored looks identical, in the response, to a rule that was loaded and followed.20
SLSean Leeinseanleecoder.hashnode.dev·Apr 22 · 14 min readRoot-default / screen-claim tracking: redesigning Expo Router analytics with two AIs in the loopWhere I picked up the MR A new requirement had come in: every Expo Router screen should fire a screen_view event, with a handful of screens attaching extra payload. I sketched an initial direction wit20
