
structure|1️⃣ 三级笔记、思想框架
OpenAI 推出 Workspace Agents——从「个人提效」迈向「组织协作」的关键一步。Agent 不再只是回答 prompt 的工具,而是能跨工具、跨团队、持续运行的 workflow 执行者。
"Build once, improve through use, then share or duplicate for new workflows."
Rippling 案例(Ankur Bhatt):
"The hard part of building an agent is not the model. It's the integrations, memory, the user experience."
concepts|2️⃣ 关键概念、概念网络
context:
"Teams can now create shared agents that handle complex tasks and long-running workflows, all while operating within the permissions and controls set by their organization."
费曼一下:OpenAI 在 ChatGPT 中推出的组织级 AI Agent。不是个人聊天助手,而是能被团队共享、在云端持续运行、跨多个工具执行复杂工作流的自动化执行者。它是 GPTs 的进化体,从「个人玩具」升级为「组织基础设施」。
context:
"Agents are powered by Codex in the cloud, giving them access to a workspace for files, code, tools, and memory."
费曼一下:Workspace Agents 的底层引擎。Codex 赋予 Agent 写代码、运行代码、调用工具的能力,让 Agent 不只是对话,而是真正的执行者。Codex 在云端为每个 Agent 提供一个包含文件、代码、工具和记忆的 workspace。
context:
"They can take on many of the tasks people already do at work—from preparing reports, to writing code, to responding to messages. They run in the cloud, so they can keep working even when you’re not."
费曼一下:长时间运行的工作流——不是一问一答的单次交互,而是可以持续运行数小时甚至数天的多步骤任务。你人已经离开,Agent 还在云端继续工作。这是 Workspace Agents 区别于传统 GPTs 的核心特征之一。
context:
"Many of the most important workflows inside an organization depend on shared context, handoffs, and decisions across teams. Workspace agents are designed for that kind of work."
费曼一下:组织内跨团队共享的信息、上下文和知识。个人 AI 助手只能解决个人任务,但组织中最重要的工作往往依赖于跨团队的共享信息、交接和决策。Workspace Agents 正是为这类场景设计的。
context:
"Because agents have memory and can be guided and corrected in conversation, they get better as teams use them."
费曼一下:Agent 的记忆能力——能记住之前的交互和学到的东西。这让 Agent 不是一次性的工具,而是能随使用持续改进的实体。团队可以在对话中引导和纠正 Agent,它会越用越好。
agentic reading|3️⃣ 费曼 x3