Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/jkh2/Primordial-Sim/llms.txt

Use this file to discover all available pages before exploring further.

Welcome to the Primordial documentation. Primordial is a continuous-space ecosystem simulator built on WebGL, featuring an integrated AI Lab Partner that observes, designs experiments, and writes research reports on the living world — no install required, no backend, no dependencies.

Quick Start

Open Primordial in your browser and explore your first ecosystem in minutes

Controls & UI

Mouse interactions, keyboard shortcuts, and panel navigation

Ecosystem Mechanics

How organisms eat, hunt, flock, reproduce, and die

Genetic Evolution

Heritable traits, mutation, and emergent natural selection

AI Lab Partner

Connect an AI provider to analyze, experiment, and report

Scenario Presets

Six tuned scenarios from peaceful aquarium to extinction event

Get started in 30 seconds

1

Open Primordial

Visit jkh2.github.io/Primordial-Sim in any modern browser with WebGL support (Chrome, Firefox, Safari, Edge). No install, no sign-up.
2

Choose a scenario

Select a preset from the Scenario Presets dropdown in the World tab — try Stable Eden for a peaceful start or Arms Race to watch evolution accelerate.
3

Interact with the world

Left-click anywhere to drop food. Right-click to spawn organisms. Hover over any dot to inspect its species, genes, and kill count. Press Space to pause.
4

Connect the AI Lab

Press L or click the star icon (top-right) to open the AI Lab Partner panel. Add your API key and click Analyze Now to get an immediate ecological report on your living world.

Why Primordial

Education

Natural selection, predator-prey dynamics, carrying capacity, and genetic drift all emerge visibly without any scripting — observable in real time

Research intuition

A fast interactive sandbox for building intuition about how parameter changes cascade through an ecosystem

AI-assisted science

The AI Lab Partner formulates hypotheses, designs controlled experiments, executes them, collects data, and produces analysis reports

Zero friction

Single HTML file, no build step, no backend — runs fully offline except for AI API calls

Build docs developers (and LLMs) love