最新资讯
dev.to 用 Switch 构建 AI 投资委员会:多空研究员 Agent 协作 bloomberg.com 微软正结束向甲骨文租用 AI 云算力的协议 the-decoder.com Perplexity 推出常驻 AI Agent“Teammates” dev.to osFoundry的Loop和Graph是什么,何时该用哪个? dev.to 限制上下文而非限制评审:面向开源补丁的字节预算证据包 dev.to 约束自主智能体:用 GKE Agent Sandbox 实现亚秒级零信任执行 dev.to 实测:给 Claude Code 十个翻译工具,三次运行中有一次完全不用 unite.ai Ramp 数据:企业 AI token 用量增速超过支出增速 pub.towardsai.net AI 生成的 ICLR 评审对被拒论文更宽松:如何压力测试你的 rebuttal blogs.nvidia.com NVIDIA 开源工具如何助力研究者构建下一代 6G 网络 wired.com AI 公司正积极收购企业工作文件与邮件用于模型训练 wechat 网友用AI搓出来的“App”,可以直接在小红书玩起来了?
多元动态
登录 注册
智能体 github.com ·

BuilderIO 开源 Agent-Native:让 agent 与 UI 共享同一 action 层的 TypeScript 框架

原文 我的收藏

Agent-Native 是 BuilderIO 开源的 TypeScript 框架,用于构建将自主工作与专用 UI 配对的 agent。其核心是“共享 action”:每个能力只定义一次,agent 将其作为工具调用,UI 则通过代码调用,二者共用同一套校验、权限与实现;同时共享数据和应用状态,agent 不通过点击 UI 操作,而是走与 UI 相同的 action 层。框架内置 agent 聊天、认证与权限、技能与记忆、自动化、agent 团队,以及 PostgreSQL(生产)和 PGlite(本地)后端,并可通过 HTTP、MCP、A2A 和 CLI 暴露 action。项目还提供 Clips、Design、Slides、Analytics、Calendar、Mail 等示例 agent。

AI推荐理由 BuilderIO 开源的 TypeScript 框架 Agent-Native,核心设计是把每个能力定义为 action,同一 action 同时作为 agent 工具和 UI 调用入口,并共享数据、权限与应用状态;还内置认证、技能/记忆、自动化、agent 团队和 PostgreSQL/PGlite 后端,并暴露 HTTP、MCP、A2A、CLI 多种接口。对构建带 UI 的 agent 应用有实际参考价值,但正文为项目自述,未提供性能、稳定性或生产验证证据,安装与星标等信号不能证明可靠性。

72 Star 6,516 榜单名次 24

正文

# Agent-Native Agent-Native is an open-source TypeScript framework for building agents that pair autonomous work with a purpose-built UI. Define each capability once as an [action](https://agent-native.com/docs/actions-overview): the agent uses it as a tool, and the UI calls it from code. ## Quick start ```bash npx --yes @agent-native/core@latest create my-agent --standalone --template chat ``` Follow the [getting started guide](https://agent-native.com/docs/getting-started) for a full intro to the framework. ## Why build agents with UIs? Coding agents work with more than a text box. Their environment provides context, tools, files, tests, and previews that make their capabilities and results visible. Knowledge work needs that same kind of environment. A UI shows what an agent can do and gives people familiar ways to inspect, edit, approve, and share its work. ## How Agent-Native works - **[Shared actions](https://agent-native.com/docs/actions-overview).** The agent calls each capability as a tool, and the UI calls it from code. Both paths use the same validation, permissions, and implementation. - **[Shared data](https://agent-native.com/docs/server-database).** Work done by the agent appears in the UI, and work done in the UI is available to the agent. - **[Shared application state](https://agent-native.com/docs/context-awareness).** The agent receives relevant UI state, such as the current page, selected record, or active view. The agent does not click through the UI. It works through the same action layer as the UI. ### Example: a shared action Create `actions/hello.ts`: ```ts import { defineAction } from "@agent-native/core/action"; import { z } from "zod"; // One action powers every app surface: UI, agent, HTTP, MCP, A2A, and CLI. export default defineAction({ description: "Return a friendly greeting.", schema: z.object({ name: z.string().default("world").describe("Name to greet"), }), http: { method: "GET" }, run: async ({ name }) => { return { message: `Hello, ${name}!` }; }, }); ``` The agent receives `hello` as a tool. React calls the same function with `useActionQuery("hello", { name: "Alex" })`. Agent-Native also exposes it through HTTP, MCP, A2A, and the CLI. ## Included - **[Agent chat](https://agent-native.com/docs/agent-surfaces):** Let people delegate work, ask questions, and review results in the same UI. - **[Authentication and permissions](https://agent-native.com/docs/authentication):** Control who can access and change shared work. - **[Skills and memory](https://agent-native.com/docs/agent-resources):** Give agents reusable expertise and persistent context. - **[Automations](https://agent-native.com/docs/automations):** Run agent work on schedules or events. - **[Agent teams](https://agent-native.com/docs/agent-teams):** Delegate work to specialist agents in the same workspace or across connected agents. - **[PostgreSQL backend](https://agent-native.com/docs/server-database):** Use PostgreSQL in production and PGlite for local development on any Nitro-compatible host. Bring your LLM, SQL database, tools, and infrastructure. Everything you build stays yours. See Agent-Native in action: https://github.com/user-attachments/assets/ef51644b-6506-46d8-8083-0af7b7e5b65c ## Open-source agents Start from one of these agents or use it as an example for your own. **Clips** Record and understand meetings, screens, and voice notes. **Design** Generate and refine interactive designs. **Slides** Create and edit on-brand presentations. **Analytics** Ask questions of your data and build dashboards. **Calendar** Find time, schedule events, and manage bookings. **Mail** Prioritize email, draft replies, and follow up. **Assets** Create and organize on-brand media. **Content** Draft, organize, and publish content. **Plans** Create and review visual plans with diagrams, wireframes, and prototypes. Explore the [full app gallery](https://agent-native.com/apps), or start with the [framework guide](https://agent-native.com/docs/getting-started). ## Documentation and community - Read the [documentation](https://agent-native.com/docs). - Join [Discord](https://discord.gg/qm82StQ2NC) to ask questions, share what you're building, and get help. ## Contributing Working on this repository itself? See [DEVELOPMENT.md](./DEVELOPMENT.md) for local setup, workspace structure, and guard scripts. ## License MIT

来源报道 · 2 篇 · 以下链接均前往原始媒体

← 返回精选动态 原文

更多内容