Docs

Why Colony uses Linux network namespaces instead of Docker containers for AI agent isolation. A technical comparison.

How Colony compares to Claude Code, OpenAI Codex, Google Antigravity, Cursor, and other agentic coding tools in 2026. Colony orchestrates Claude Code sessions with isolation — it doesn't replace the agent.

Jujutsu (jj) eliminates merge conflicts between parallel AI agents. Here's how it works and why we chose it.

How Colony enables multiple AI agents to work simultaneously with kernel-level isolation and zero container overhead

Real-time preview of agent work through web interfaces, terminals, and streaming activity logs in your browser

Full lifecycle from spawn to stop — states, transitions, supervision trees, and service orchestration in Colony

Real-time web interface for managing colonies with live previews, terminal access, and agent monitoring.

Configure colony environments with colony.toml — define services, ports, environment setup, and provisioning.

Connect AI coding agents to colonies — HumanLayer hld daemon integration, streaming events, and brood orchestration.

Colony's Functional Source License (FSL) — source-available today, open source tomorrow.

Colony development roadmap — what we're building and when.

Common questions about Colony — supported agents, pricing, licensing, platform requirements, and more.

How Colony uses Nix Flakes for reproducible development environments with pinned dependencies.

Complete reference for colony.toml, environment variables, and runtime configuration options.

Deep dive into Stem's Rust + Ratatui architecture, IPC communication, and keyboard-driven UX.

How Colony uses Linux network namespaces, Caddy reverse proxy, and dnsmasq for isolated colony networking.

Understanding Colony's security guarantees, isolation boundaries, and threat model.