Inside Claude Code’s Leak: 8 Compaction Modes, 3 Memory Tiers, 44 Flags Anthropic Never Talked…
中文摘要
泄露的Claude Code源码揭示了Anthropic未公开的压缩模式、内存层级及44个标志位，展示了AI智能体处理长对话的底层技术实现。
English Summary
Leaked Claude Code source code reveals Anthropic's hidden compaction modes, memory tiers, and flags, detailing how production agents maintain persistence during long conversations.
Original Excerpt
What 512,000 lines of accidentally-published TypeScript actually reveal about how a production agent stays alive across long conversations. Continue reading on Towards AI »