aiming-lab/AutoResearchClaw：README 來源編輯指南

aiming-lab/AutoResearchClaw 的 README 將專案描述為「Fully autonomous & self-evolving research from idea to paper. Chat an Idea. Get a Paper. 🦞」。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「README」下寫到：📄 Our paper is on arXiv , come read it! AutoResearchClaw: Self-Reinforcing Autonomous Research with Human-AI Collaboration。這說明的是專案邊界，不是已完成的生產驗證。

從 README 的「🔥 News」與相關條目，可以先判斷它是否處理你的實際問題：[04/01/2026] v0.4.0 , Human-in-the-Loop Co-Pilot System , AutoResearchClaw is no longer purely autonomous. New HITL system adds 6 intervention modes (full-auto, gate-only, checkpoint, step-by-step, co-pilot, custom), per-stage policies,。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：[05/19/2026] v0.5.0 , Multi-Domain Experiment Agents + ARC-Bench , Two headline updates. (1) Domain-specialist execution agents: the experiment stage (Stages 10,13) now routes beyond the default ML sandbox to specialist agents per field ,。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「🤔 What Is This?」等段落。可確認的線索包括：You think it. AutoResearchClaw writes it. You guide the key decisions.。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： # Fully autonomous , no human intervention pip install -e . && researchclaw setup && researchclaw init && researchclaw run --topic "Your research idea here" --auto-approve # Co-Pilot mode , collaborate with AI at key decision points researchclaw run --topic "Your research idea here" --mode co-pilot 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「⚡ One Command. One Paper.」，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「🤔 What Is This?」段落提到：Drop a research topic , get back a full academic paper with real literature from OpenAlex, Semantic Scholar & arXiv, hardware-aware sandbox experiments (GPU/MPS/CPU auto-detected), statistical analysis, multi-agent peer review, and。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：[03/30/2026] Flexible Skill Loading , AutoResearchClaw now supports loading open-source and custom skills from any discipline to further enhance your research experience.。