Claude Code From Scratch reverse-engineers Anthropic’s coding AI with a minimal perception-action loop and typed tool dispatch. This repo offers a clear, incremental path to understanding a production-ready AI harness.
learn-harness-engineering offers a practical 5-subsystem harness framework to improve AI coding agent reliability, backed by industry research and starter templates for Claude Code and Codex.
Symphony by OpenAI orchestrates autonomous coding agents via work boards and proof-of-work validation, shifting AI coding from direct supervision to task-level management.
