What Broke Today

Hey PH! 👋 Built this as a weekend project that mass-enabled Telegram alerts as a distraction from real work—then realized it could actually be useful. The problem: Status pages are corporate poetry. "We're experiencing elevated error rates" = "everything is on fire and the on-call engineer is crying." So I made a monitor that: - Watches 100+ service status pages via RSS - Translates corporate speak into snarky, actually-readable updates - Sends alerts faster than your coworker's "is Slack down?" message The fun part: Every incident summary is AI-generated. Claude reads the boring status update and writes what it actually means. "Degraded performance in us-east-1" becomes "AWS decided data persistence was optional again." Built the whole thing with Claude Code in about a day—Python backend, FastAPI, SQLite, and a terminal aesthetic because I have mass-enabled Telegram as a distraction from real work. Tech stack: 🐍 Python + FastAPI + SQLite 🤖 Claude for the snarky commentary 🖥️ Claude Code for rapid development 📡 RSS monitoring + Telegram alerts Would love feedback: - What services should I add? - What would make this useful for your on-call rotation? - Too snarky or not snarky enough? And yes, if our monitoring goes down, who monitors the monitors? It's turtles all the way down. 🐢