Skip to main content
Documentation Glossary

Glossary

Key terms and concepts in the Ants at Work colony

Core Concepts

Ant
A digital agent in the colony that follows simple rules leading to emergent behavior. Ants communicate through pheromones, not direct messages.
Colony
A group of ants working together through stigmergy. The top-level organizational unit that coordinates distributed intelligence.
Stigmergy
Coordination through environmental modification. Ants communicate by depositing and sensing pheromones rather than direct messaging.
Emergence
Complex behavior arising from simple rules. The colony is intelligent; individual ants are not. Intelligence emerges from collective behavior.
Pheromone
A signal deposited on edges in the graph. Higher pheromone levels make paths more attractive to other ants.

Ant Castes

Caste
A specialized role for ants. Each caste has different pheromone sensitivity and behavior patterns.
Scout
An ant caste with low pheromone sensitivity (0.3). Scouts explore new paths and discover unknown territory.
Harvester
An ant caste with high pheromone sensitivity (0.9). Harvesters exploit known, proven paths.
Relay
An ant caste with medium pheromone sensitivity. Relays spread information between different regions of the colony.
Hybrid
An ant caste with adaptive pheromone sensitivity. Hybrids switch between exploration and exploitation based on context.
Hunter
Specialized ants for the Hunt BTC mission. Includes Tame and Wild varieties that work together using Pollard's Kangaroo algorithm.

Navigation & Paths

STAN
Stigmergic A* Navigation. The core pathfinding algorithm that combines A* search with pheromone-based guidance.
Effective Cost
The STAN formula: base_weight / (1 + pheromone x sensitivity). Lower effective cost makes edges more attractive.
Edge
A connection between nodes in the graph. Edges carry pheromone levels and base weights.
Node
A point in the graph. Can represent a concept, region, market state, or computational position.
Trail
A sequence of edges with elevated pheromone levels. Trails represent discovered paths through the problem space.
Superhighway
An edge with pheromone level greater than 20. Represents crystallized, proven knowledge that persists.

Pheromone Dynamics

Deposit
The act of leaving pheromone on an edge after successful traversal. Reinforces good paths.
Decay
The gradual evaporation of pheromone over time. Prevents lock-in to suboptimal paths and allows adaptation.
Pheromone Sensitivity
How strongly an ant responds to pheromone (0.0-1.0). Low sensitivity favors exploration; high sensitivity favors exploitation.
Crystallization
When temporary pheromone patterns become permanent knowledge. Superhighways and patterns are crystallized learnings.

Organization

Queen
The colony coordinator. Manages caste allocation, mission strategy, and overall colony health.
Mission
A specific problem the colony is solving. Examples include Hunt BTC (puzzle solving) and Trade (market analysis).
Group
An organizational container (Colony, Mission, Team). Provides isolation and permission boundaries.
Actor
An entity that can perform actions. Includes humans, AI agents, and system processes.
ONE Ontology
The 6-dimension framework: Groups, Actors, Things, Connections, Events, Knowledge. The universal schema for all colony data.

Infrastructure

Gateway
The Cloudflare Worker API at api.ants-at-work.com. Provides fast coordination for distributed workers.
TypeDB
The graph database that stores the colony's environment, events, and knowledge. Single source of truth.
Worker
A distributed compute node running ants-worker. Contributes processing power to colony search.
Growth Gate
Security checkpoint limiting colony size. The colony must pass requirements before scaling up.
Kill Switch
Emergency mechanism to halt colony operations if anomalies are detected. Armed at all times.

Hunt BTC Mission

Kangaroo Algorithm
Pollard's Kangaroo method for solving the discrete logarithm problem. Used in the Hunt BTC mission.
Tame Hunter
A hunter ant that starts from known points and walks forward deterministically. Part of the Kangaroo algorithm.
Wild Hunter
A hunter ant that starts from the target and walks forward. Collision with a Tame Hunter recovers the key.
Distinguished Point
A rare landmark in Pollard's Kangaroo algorithm. When both tame and wild kangaroos land on the same DP, collision is detected.
Collision
When a tame and wild kangaroo land on the same distinguished point. This allows recovery of the private key.
Region
A subdivision of the search space. Regions track pheromone levels and worker activity.

Genetics & Evolution

Genome
Heritable parameters that define ant behavior. Genomes can mutate during reproduction, enabling evolution.
Fitness
A measure of how well an ant's genome performs. High-fitness genomes reproduce; low-fitness ones die out.
Role
Permission level in the colony hierarchy: platform_owner > group_owner > operator > worker > viewer.