garrytan/gbrain：README 來源編輯指南

garrytan/gbrain 的 README 將專案描述為「Garry's Opinionated OpenClaw/Hermes Agent Brain」。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「GBrain」下寫到：Search gives you raw pages. GBrain gives you the answer. It's the brain layer your AI agent has been missing , the only one that does synthesis, graph traversal, and gap analysis in one box.。這說明的是專案邊界，不是已完成的生產驗證。

從 README 的「GBrain」與相關條目，可以先判斷它是否處理你的實際問題：A self-wiring knowledge graph. Every page write extracts entity refs and creates typed edges (attended, worksat, investedin, founded, advises) with zero LLM calls.。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：A synthesis layer that gives you the actual answer. Synthesized, well-cited prose across people, companies, deals, and ideas. Not "here are 10 chunks that mention your query";。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「GBrain」等段落。可確認的線索包括：And now it works as a company brain too. Each person on the team gets their own slice of the brain, scoped by login. When you query, you only see what you're allowed to see , never another person's notes, never another team's data.。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： gbrain init --pglite # 2-second local brain (no Docker) claude mcp add gbrain -- gbrain serve # or: codex mcp add gbrain -- gbrain serve 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「What this looks like」，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「GBrain」段落提到：Lots of personal-knowledge systems give you keyword matching and grep in a box. GBrain does that, and adds two things nobody else ships together:。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：Codex. Codex reads the bearer from $GBRAINREMOTETOKEN at runtime, so the token never lands in Codex config.。