Claude Code Reverse-Engineered: 98.4% of Codebase is Operational Harness

A reverse-engineering analysis of Anthropic's Claude Code by UCL researchers reveals that only 1.6% of its codebase is AI decision logic, with the remaining 98.4% dedicated to operational infrastructure. This finding suggests that a robust deterministic harness is the primary differentiator for AI agents as frontier models converge in raw ability.