Agent Harness 是什么：两种构建生产级智能体的方式
原标题：Article: The Agent Harness: What It Is and Two Ways to Build One
AI 摘要
文章提出「Agent Harness」概念，认为生产级智能体等于模型加外围工程层，该层分为开发（记忆、工具与MCP、检索、编排）和运维（可观测性、评估、护栏、路由、成本监控、部署扩展）两半。作者对比了 Harness-as-a-Service（如 AWS AgentCore、Google Vertex AI Agent Engine、Azure AI Foundry Agent Service、LangGraph Platform）与自管理方案（LangChain/LlamaIndex 加 Agent Router 或 LiteLLM 部署在 Kubernetes），指出两者能力相同，区别在于由谁运维、成本结构与控制权取舍。结论是应以最小化起步、随智能体演进逐步扩展 harness，避免过度工程。
正文节选
Key Takeaways - The gap between a demo and a production agent is the harness: Everything you build around the model to make it a real product. Almost none of it comes from the model itself, even though it is where most of your engineering time actually goes. - The harness splits into two halves: Development extends what the model can do, while operations keeps it running once real users show up. That operations half is mostly DevOps in a new hat. - Harness-as-a-Service (HaaS) (like AWS AgentCore