[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tag-agent-harness":3},{"tag":4,"articles":11},{"id":5,"name":6,"slug":7,"article_count":8,"description_zh":9,"description_en":10},"71509a7d-198a-410c-98cd-8ef806cc476b","agent harness","agent-harness",4,"Agent harness 指的是包住 LLM agent 的外層系統：工具呼叫、追蹤、評測、流程控制與失敗回復。對要把 agent 放進產品的團隊來說，重點不只模型能力，而是能否穩定、可觀測地運作。","An agent harness is the layer around an LLM agent that handles tool calls, tracing, evals, workflow control, and recovery. For product teams, reliability depends less on the model alone and more on this surrounding system.",[12,21],{"id":13,"slug":14,"title":15,"summary":16,"category":17,"image_url":18,"cover_image":18,"language":19,"created_at":20},"574953d9-dafe-4fd3-b4da-133f2ed9f2c9","hermes-agent-agent-harness-framework-en","Hermes Agent: The Agent Harness Framework to Watch","Hermes Agent aims to make agent testing and orchestration easier, with tool use, evals, and workflow control in one stack.","ai-agent","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775207577974-56eu.png","en","2026-04-03T09:12:33.141965+00:00",{"id":22,"slug":23,"title":24,"summary":25,"category":26,"image_url":27,"cover_image":27,"language":19,"created_at":28},"a9b22aa6-768c-44b1-967a-1b4ea3c28ce9","agent-harness-ai-engineering-2026-en","Agent Harness Is Quietly Defining AI Engineering","Martin Fowler, Anthropic, and OpenAI are all pointing to the same idea: agent reliability depends on the system around the model.","industry","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775057900119-b2du.png","2026-04-01T10:15:35.076717+00:00"]