Decides hand-vs-compiler for loop transforms (unrolling, SIMD, fusion, hoisting). Use when reviewing/authoring a hot loop or tempted to hand-optimize one.
Improves code quality across duplication, efficiency, and architectural fit. Use when code passes tests but quality is poor or before a major release.
Build a project skill library in .claude/skills/ via discovery, parallel authoring, and review. Use when packaging tribal knowledge. Do not use for one skill.
Explain plugin, skill, command, agent, and hook mechanics used here. Use when authoring or debugging plugins. Do not use for ops; use night-market-operations.
States load-bearing decisions, invariants, and weak points. Use when judging a design change. Do not use for gating; use night-market-change-control.
Rebuild the dev environment: uv, Python tiers, pins, traps. Use when onboarding or toolchain breaks. Do not use for daily commands; use night-market-operations.
Classify, gate, and review changes. Use when landing a PR, releasing, or amending rules. Do not use for failure triage; use night-market-debugging-playbook.
Search and record project memory (Discussions, journal, ADRs). Use before re-investigating anything. Do not use for settled battles; see failure-archaeology.