systematic-debugging

# Systematic Debugging Based on the obra/superpowers debugging methodology. ## Core Principle **NO FIXES WITHOUT ROOT CAUSE INVESTIGATION FIRST** Systematic debugging is FASTER than guess-and-check thrashing. ## Four Essential Phases ### Phase 1: Root Cause Investigation **Step 1: Read the error carefully** ```bash # Full error with stack trace (example command — use your stack's dev/run command) npm run dev 2>&1 | head -50 ``` **Step 2: Reproduce consistently (use the Reproduction Template below)** Fill it in before proceeding — if you can't reproduce, you can't fix: ````markdown ## Reproduction Report **Bug one-liner:** [what breaks in 1 sentence] **First seen:** [date/commit if known] **Severity:** [blocker / critical / high / medium / low] ### Environment - **Where:** [dev local / preview deploy / production / CI] - **Browser/device:** [Chrome 123 desktop / Safari iOS / any — if applicable] - **Auth state:** [logged in as role X / logged out / new user] - **Data state:** [empty DB / dataset with N records / specific fixture] ### Steps to reproduce 1. [precise step — URL + action] 2. [precise step] 3. [observe] ### Expected [what should happen] ### Actual [what happens + error message / screenshot path] ### Reliability - [ ] Always reproduces (100%) - [ ] Intermittent — frequency: [1 in 5 / 1 in 20 / random] - [ ] Only on specific data / timing / race ### Console / network / logs ```text [paste error, stack trace, failed request] ``` ### Minimal re...