已核验 · Aug 5, 2026
已独立佐证Productivity agents 集群:ChatGPT Scheduled Tasks + Claude Skills + Vercel AI Gateway + Notion Agents + Linear AI + Figma Make 一周全 GA
8 个信源2026-08-02 这一周出现了一组 productivity tool GA 集群,都发货 agent 面:OpenAI ChatGPT Scheduled Tasks(recurring 模型调用)、Anthropic Claude Skills(按需程序化工作流)、Vercel AI Gateway(多 provider 路由)、Notion AI Agents(工作区常驻工作流)、Linear AI(issue triage)、Figma Make(prompt-to-design primitives)。The Decoder 把这个集群框成「每个 productivity tool 都发货 agent 面」。模式:每个 tool 发货的 agent 住在 tool 现有数据模型里,scope 到 tool 的 primitives,跑在 trigger 上(手动、定时、事件)。
为什么现在讲
这个集群是把六条独立 GA 变成一个连贯「productivity agent 面」故事的 editorial framing —— 并排对比对创作者最有用。
推荐理由
演示空间:同一创作者工作流(每周 recurring brief)在 Scheduled Task + Skill + AI Gateway + Notion Agent + Linear triage + Figma Make 原型上的并排对比视频。
依据
The Decoder + IT 之家周报 + 六个独立 vendor 一手源
“六个 productivity tool 在同一周发了 agent 面 —— ChatGPT Scheduled Tasks、Claude Skills、Vercel AI Gateway、Notion AI Agents、Linear AI、Figma Make —— 模式(agent scope 到 tool 现有 primitives)是故事本身。”
切入角度
用集群引入「productivity-tool agent 面」模式 —— 每个 tool 发货一个住在现有数据模型里的 agent —— 用这个 lens 并排对比 vendor 方案。
形式
长视频讲解
演示想法
录一段 16 分钟并排对比讲解:2 分钟开场讲「productivity-tool agent 面模式」,然后每个 tool 2 分钟(Scheduled Tasks / Skills / Gateway / Notion / Linear / Figma),最后 4 分钟在同一 recurring weekly-brief 工作流上并排对比六个。
平台注意
每个 vendor GA 都把发布框成跟自己关心的竞争集对比;The Decoder 和 IT 之家是 editorial framing 层,不是独立验证。记录任何具体能力声明前,对照 vendor 一手文档确认。
可用说法
- OpenAI promoted Scheduled Tasks for ChatGPT to GA, exposing a Tasks API for programmatic management of recurring and one-shot model calls.
- Anthropic promoted Claude Skills to GA — folder-bundled procedural knowledge (SKILL.md + scripts + references) loaded on demand via the tool_search tool.
- Vercel promoted AI Gateway to GA — a unified LLM router with multi-provider fallback, retry, spend limits, and per-request cost attribution.
- Notion promoted Notion AI Agents to GA — workspace-resident agents that read, write, and trigger workflows on a trigger (manual, scheduled, or event).
- Linear shipped AI-powered issue triage — automated project / label / assignee suggestion, duplicate detection, and AI-generated summaries, positioned as first-pass triage.
- Figma promoted Figma Make to GA — a prompt-to-design prototype generator that produces editable Figma primitives (frames, components, auto-layout) from a natural-language description.
证据链
来自新闻
- The Decoder:科技媒体报道「productivity agents」GA 集群
- IT 之家:中文媒体覆盖 8/2 productivity agents 集群
- ChatGPT Scheduled Tasks 正式 GA,Tasks API 支持编程式调度
- Anthropic Claude Skills 正式 GA —— 文件夹打包的程序化知识
- Vercel AI Gateway 正式 GA —— 统一 LLM 路由 + fallback + 单请求成本归因
- Notion AI Agents 正式 GA —— 工作区常驻 agent,读、写、触发工作流
- Linear AI issue triage —— 自动化打标、重复检测、AI summary
- Figma Make 正式 GA —— prompt-to-design 原型生成器,产出可编辑 Figma primitives
拆解
六个 productivity tool 同一周发了 agent 面 —— editorial framing(「每个 productivity tool 都发货 agent」)有用,但如果你不引入底层模式,内容就退化成 vendor directory。本篇解释怎么用集群引入「productivity-tool agent 面」模式(agent scope 到 tool 现有 primitives,跑在 trigger 上),并用这个 lens 并排对比 vendor 方案。
信源
- The Decoder: tech-press coverage of the 'productivity agents' GA cluster for the week of 2026-08-02
- IT之家: 中文科技媒体覆盖 8/2 productivity agents GA 集群
- OpenAI: ChatGPT Scheduled Tasks GA + Tasks API
- Anthropic: Claude Skills GA — procedural skill bundles for repeated workflows
- Vercel: AI Gateway GA — unified LLM routing, fallbacks, cost analytics
- Notion: Notion AI Agents GA — autonomous workflow agents in the workspace
- Linear: Linear AI for issue triage — automated labeling and routing
- Figma: Figma Make GA — prompt-to-design prototype generator
风险
- Use The Decoder and IT之家 as media-type corroboration, but read the underlying vendor docs for any specific capability claim before stating it on the record. Verify specific capability claim against the underlying vendor docs and the actual license / pricing matrix before stating it on the record; do not paraphrase per-platform pricing or license terms into specific dollar figures or commercial-use clauses.
- OpenAI platform docs confirm Scheduled Tasks GA but the exact per-task rate is not in this pass. Verify specific capability claim against the underlying vendor docs and the actual license / pricing matrix before stating it on the record; do not paraphrase per-platform pricing or license terms into specific dollar figures or commercial-use clauses.
- Anthropic Skills docs confirm GA but the pricing tier is not in this pass. Verify specific capability claim against the underlying vendor docs and the actual license / pricing matrix before stating it on the record; do not paraphrase per-platform pricing or license terms into specific dollar figures or commercial-use clauses.
- Linear docs explicitly call out that the human reviewer remains in the loop. Verify specific capability claim against the underlying vendor docs and the actual license / pricing matrix before stating it on the record; do not paraphrase per-platform pricing or license terms into specific dollar figures or commercial-use clauses.
演示思路
- 并排 weekly-brief 工作流:在 Scheduled Tasks + Skills + Gateway + Notion + Linear + Figma Make 上跑同一 recurring brief,画 time-to-first-output
- 决策树:「哪个 tool 面配哪个工作流」(recurring 模型调用 → Scheduled Tasks,repeated 程序化工作流 → Skills,多 provider 路由 → AI Gateway,工作区 agent → Notion,issue triage → Linear,设计 primitives → Figma Make)
- 时间线图:把每个 vendor GA 放在同一时间线,展示集群