Why Learn Claude?
Why this project exists, what it teaches, and how it fits into the LIGHT HOPE ecosystem.
What Changed in AI Coding Systems
Claude Code is not a chat interface with code output. It is a terminal-native agent loop with persistent memory, lifecycle hooks, reusable skills, subagent delegation, and autonomous background workflows. This represents a fundamental shift from "AI assistant" to "AI system."
What Claude Code Represents
512,000 lines of production TypeScript. 40+ plugin tools. 44 feature flags for unreleased capabilities. 5 context compaction strategies. A multi-agent coordinator. This is what a serious agentic coding platform looks like internally — and it took years to build.
What LightHope Can Learn
Agent loops, project memory, hooks and automation points, skills and reusable workflows, subagents and role specialization, always-on background workflows, productized coding systems. Many of these map directly to existing LIGHT HOPE ecosystem directions: voice persistence, project governance, nightly operations, multi-agent role decomposition.
Boundary: What We Study vs. What We Don't
We study engineering patterns — not leaked code. We reference official documentation and public reporting — not unauthorized archives. We build internal abstractions — not mirrors. This boundary is mandatory and non-negotiable.