yvgude/lean-ctx：README 來源編輯指南

yvgude/lean-ctx 的 README 將專案描述為「Control what your AI can see. LeanCTX (Lean Context) is the context intelligence layer for AI agents , one local Rust binary that decides what they read, remembers what they learn, guards what they touch, and proves what they save.」。本文只整理倉庫可直接核對的內容，不把 star、Fork 或宣傳語當成品質證明。README 在「Control what your AI can see.」下寫到：LeanCTX , Context Engineering Layer for AI Coding Agents。這說明的是專案邊界，不是已完成的生產驗證。

從 README 的「Why developers use LeanCTX」與相關條目，可以先判斷它是否處理你的實際問題：Lower API costs , 60,90% fewer tokens on reads and shell output, cached re-reads cost 13 tokens。若需求不同，不應只因專案熱度就採用。本文保留原始專案名、命令與元件名，方便回到一手來源核對。 README 另外列出一項可核對的資訊：Longer useful coding sessions , less context waste = more room for actual code reasoning。這類原文條目可用來設計試跑步驟，但不能取代實際環境測試。

README 將運作方式分散在「Control what your AI can see.」等段落。可確認的線索包括：| Problem | With LeanCTX | |---------|-------------| | Repeated file reads: 2000 tokens each | Cached re-reads: 13 tokens | | Raw git status: 800 tokens | Compressed: 120 tokens | | Every turn re-sends the whole history | Proxy compresses。本文不把未寫出的架構、效能或安全邊界補成結論；真正的執行鏈仍要配合目錄、設定檔與版本標籤檢查。

第一次安裝應從 README 指出的入口開始。目前可核對的命令是： # 1) Install (pick one) curl -fsSL https://leanctx.com/install.sh | sh # universal (no Rust needed) brew tap yvgude/lean-ctx && brew install lean-ctx # macOS / Linux npm install -g lean-ctx-bin # Node.js cargo install lean-ctx # Rust pi install npm:pi-lean-ctx # Pi Coding Agent # 2) One-command setup for your agent lean-ctx wrap cursor # or: wrap claude / wrap codex / wrap vscode # Done. Savings appea 如果倉庫沒有命令，本文不會自行編造步驟，而是建議先閱讀「Why developers use LeanCTX」，確認系統依賴、預設埠與首次初始化。

日常使用取決於專案文件。README 的「Control what your AI can see.」段落提到：> Control what your AI can see. LeanCTX , short for Lean Context , is the context engineering layer for AI coding agents: one local Rust binary that helps agents read repositories, run development commands, compress context sent to the。設定檔、環境變數、權限與資料目錄只在來源明確時才會記錄；沒有寫出的預設值，應在測試環境驗證並保留回滾副本。 同一部分也提到：No more "I already showed you this file" , session memory persists across chats。