Alibaba Cloud Open-Sources "AI Agent Handbook," an Enterprise-Grade Full Lifecycle Handbook for Agents
English text is being prepared.
Alibaba Cloud has open-sourced the "AI Agent Handbook," covering five parts—architecture, building, running, governance, and tuning—with Harness at its core to move enterprise Agents from pilot to governable production. The 2026 survey cited in the handbook shows that about 46% of enterprises have completed or are in the process of development, while only about 18% have actually deployed to production; about 90% have context and memory needs, about 63% have multi-model routing and cost governance needs, and fewer than about 8% have trajectory-based automated evaluation. Those paying attention to or implementing MCP total about 37%, and only about 9% have private deployments within the enterprise.
Why read this
AI score 66
The handbook lays out the architecture, governance, and tuning path for enterprise Agents moving from pilot to production, and includes 2026 survey data for reference.
Topics
Body
阿里云开源《AI Agent Handbook》:用 Harness 把企业 Agent 从试点推到可治理生产
企业级 Agent 全生命周期实践手册:架构、构建、运行、治理、调优,外加调研与案例,开源地址: https://t.co/CzC395KiJq
手册为何重写? 前言归纳三大新挑战:让概率智能变成能扛关键任务的可靠生产力;在稳定、安全、性能、成本上做成可大规模部署的基础设施;让 Agent 进入核心业务流程,而不是停在孤岛试点。写作目标还包括统一术语(Harness、Runtime、Agent、Workflow 等),减少「同一词各说各话」被模型放大后的混乱。
2026 年调研结果 · 投入成共识:已开发完成或开发中合计约 46%(去年约 36%);真正部署到生产约 18%;持续迭代更低。 形态共存:单 Agent 约 40%、多 Agent 约 42%、以 HITL 为主要架构约 15%——自主度是可调参数,不是新旧替代。 编码最先规模化:采购 Coding 工具约 57%,通用 Agent 框架约 78%,近四成两者并用;市场碎片化,可迁移工程抽象比押单一厂商更稳。 构建后难点:上下文/记忆需求约 90%;多模型路由与成本治理约 63%;评估上人工抽查约 55%,基于轨迹自动评估不足约 8%。 成功率:达 90% 以上约一成;约 16% 根本没量化过成功率。有评估体系的团队,高成功率比例明显高于无评估者(文中对比)。 MCP:关注或落地合计约 37%,企业内私有部署仅约 9%;缺的是注册中心、身份鉴权、版本灰度、审计与网关,不是「懂不懂协议」。 成熟度 L1–L4:辅助生成 → 受控自动化 → Agentic Execution → Managed & Optimizing;全书后半部基本在回答如何从 L2 往 L3/L4 走。
手册整体结构 · 架构(1–2):Agentic Application 边界、企业成熟度、参考架构(组件 / 平台责任 / 生命周期) · 构建(3–6):以 Harness 为核心:范式与责任边界;任务编排与长程推进;上下文、状态、Memory/Knowledge/Skill;受控行动(Function Calling、MCP、A2A、HITL、验证闭环) · 运行(7–12):Runtime 与沙箱;状态存储与语义资产;AI 网关统一 LLM/MCP/Agent 流量;异步与自动化;Multi-Agent;分布式通信 · 治理(13–16):可观测与成本归因;安全(注入、鉴权、高危授权、出域);Prompt/Skill/MCP/Agent 资产管理;上线前仿真与质量验证 · 调优(17–24):模型侧(SFT、Agentic RL、蒸馏等)与 Agent 侧数据飞轮:Trace→Trajectory、黄金数据集、Badcase、受控自进化、边缘运行时 · 实践(25–29 + 专题):研发效能、设计工程、运维与企业 IT、客户销售运营、GOAI Agent Infra;另有 Jev 决策模型专题、Personal Agent / Agent as a Service 专题 · 展望(30):从 Agentic Application 走向 Agentic OS