智能體交易：邁向 ACID 相容的智能體系統

摘要

大型語言模型（LLM）代理正從對話式助理演進為透過推理、工具使用、程式碼生成與工作區操作來執行長時程任務的自主系統。隨著代理日益在持久化環境與多步驟工作流程中運作，它們面臨著與交易式資料庫系統所處理之問題類似的挑戰：可靠執行、一致結果、安全並發與持久狀態管理。我們引入代理交易（agentic transaction）的概念，並提出一個符合ACID規範的代理系統框架，透過四種語意保證重新詮釋經典的ACID屬性於代理執行：語意原子性（Semantic Atomicity）、語意一致性（Semantic Consistency）、語意隔離性（Semantic Isolation）與語意持久性（Semantic Durability）。這些屬性共同為在模型不確定性與動態執行環境下建構可靠的代理系統提供了原則性的基礎。為具體實現此框架，我們開發了一個符合ACID規範的資料代理，透過交易式探索-執行-驗證循環、交易式技能中樞、基於信心分歧的驗證、語意相依性感知的隔離，以及交易感知的語意狀態管理來實現這些保證。在廣泛使用的基準測試上的實驗結果顯示，我們的系統比最先進的代理（包括Claude Code）提升了10.6%。這項工作開啟了將交易原則與系統架構延伸至建構可信、可擴展且自我進化的AI代理系統的更廣泛研究議程。

English

Large language model (LLM) agents are evolving from conversational assistants into autonomous systems that execute long-horizon tasks through reasoning, tool use, code generation, and workspace manipulation. As agents increasingly operate over persistent environments and multi-step workflows, they face challenges analogous to those addressed by transactional database systems: reliable execution, consistent outcomes, safe concurrency, and durable state management. We introduce the concept of an agentic transaction and propose an ACID-compliant agent system framework that reinterprets the classical ACID properties for agent execution through four semantic guarantees: Semantic Atomicity, Semantic Consistency, Semantic Isolation, and Semantic Durability. Together, these properties provide a principled foundation for building reliable agent systems despite model uncertainty and dynamic execution environments. To instantiate this framework, we develop an ACID-compliant data agent that realizes these guarantees through transactional exploration-execution-validation cycles, transactional skill hubs, confidence divergence-based validation, semantic dependency-aware isolation, and transaction-aware semantic state management. Experimental results on widely used benchmarks show that our system achieves a 10.6% improvement over state-of-the-art agents, including Claude Code. This work opens a broader research agenda on extending transactional principles and system architectures toward building trustworthy, scalable, and self-evolving AI agent systems.