Four AI coding harnesses all claim multi-agent, but their architectures diverge radically

都是 Multi-Agent，各家 AI 编程 Harness 的基因与信仰到底有何不同？

This piece dissects the multi-agent architectures of Claude Code, OpenAI Codex, Cursor, and Antigravity. Claude Code explores tree-based spawning and peer-to-peer Agent Teams with a shared tasks.md ledger. Codex assigns different models and reasoning effort (low/medium/high) per sub-agent to optimize cost and throughput. Cursor binds agent loops directly to IDE state, using Merkle Tree indexing and SQLite for non-blocking background edits. Antigravity enforces explicit planning with a Proceed Gate and isolates sub-agents via Git Worktree. The choice depends on whether you prioritize communication topology, compute efficiency, editing UX, or audit-grade governance.

Why it matters: A cross-sectional deep dive into four major AI coding tools' multi-agent architectures, with source-level details like shared ledgers and reasoning-affinity matching. The density is well above typical reviews. The slight discount is because it's an independent blog rather than...