Self-Driving AgentsGitHub →

Game Development

game-development

20 knowledge files6 sub-agents3 mental models

Extract gameplay design decisions, technical-art and engine choices, multiplayer/networking approaches, performance budgets, audio direction, narrative beats, and playtest feedback across Unity, Unreal, Godot, Roblox, and Blender pipelines.

Game PillarsTech Stack & PipelinesPlaytest Learnings

Install

Pick the harness that matches where you'll chat with the agent. Need details? See the harness pages.

npx @vectorize-io/self-driving-agents install game-development --harness claude-code

Memory bank

How this agent thinks about its own memory.

Observations mission

Observations are stable facts about the game's pillars, target platforms, engine and toolchain, performance budgets, content pipelines, and recurring playtest feedback. Ignore one-off bugs and discarded prototype branches.

Retain mission

Extract gameplay design decisions, technical-art and engine choices, multiplayer/networking approaches, performance budgets, audio direction, narrative beats, and playtest feedback across Unity, Unreal, Godot, Roblox, and Blender pipelines.

Mental models

Game Pillars

game-pillars

What are the game's design pillars, core loops, target audience, and platform constraints? What aesthetic and tone decisions are locked?

Tech Stack & Pipelines

tech-stack

What engine, tools, and pipelines are in use? Include performance budgets (frame time, memory, draw calls), networking model, and content authoring conventions.

Playtest Learnings

playtest-learnings

What patterns have emerged from playtests and metrics? Which mechanics, levels, and narrative beats land well or poorly, and what changes have we tried?