context.md - loop-benchmarking - Controlled experiments across agentic coding configurations. Same task, one variable, what actually works.

context.md (367B)

      1 # Project Rules
      2 
      3 - Use semantic HTML where possible
      4 - Ensure keyboard accessibility for all interactive elements
      5 - Keep the bundle small -- no heavy frameworks unless necessary
      6 - Use canvas or DOM-based rendering (either is acceptable)
      7 - Include a visible score display and next-piece preview
      8 - The game should be playable immediately on page load without user setup

	loop-benchmarking Controlled experiments across agentic coding configurations. Same task, one variable, what actually works.
	git clone https://git.shiptheloop.com/loop-benchmarking.git
	Log \| Files \| Refs \| README