跳到正文

#Agent

今日 0 条
10月5日周一
  1. elvis40

    作者受 Karpathy 关于 LLM 输出形式的讨论启发,实验一种连接高层智能体与类 Notion 灵活页面的通用界面,支持嵌入 artifacts、文本、视频、可视化讲解及 LLM 可生成的各类内容。该界面可按任务智能输出 artifacts 而非默认文本,覆盖代码审查、笔记、写作、研究、原型与模型设计等场景,降低人机协作摩擦,但作者称尚未找到完美方案。

    引用Andrej Karpathy@karpathy

    We'll be spending a lot more time trying to understand the outputs of language models. A few thoughts, tips & tricks: Writing. Something I've had success with: Ask your LLM to explain something in ASD-STE100, it's a controlled language specification originally developed for aerospace maintenance documentation. LLMs well-versed in this language and it comes with heavy constraints on clean writing style that I often find a lot more readable. Sometimes I've tried to soften it a bit e.g. ask for "80% of the way to ASD-STE100" because the spec is quite stringent. But even better: Diagrams / images. Instead of writing, ask your LLM to create a diagram. These can be a lot easier to process, parse, and understand. But even better: Web pages. Ask for output "in HTML" to get a beautiful, interactive webpage. LLMs are getting really good at frontend and can create beautiful experiences, animations, etc. But even better: Explainer videos. The output format I am most bullish on is fully custom / bespoke explainer videos generated on any arbitrary topic. Experiment with things like "Create a 3b1b style video explainer on X. Use my ElevenLabs API key for audio narration". (you'd need an API key for the latter or you can ask your LLM to find you decent free alternatives that use your local compute). This is actually starting to work! In summary: - As LLMs get better, they will do more and more of the legwork autonomously, and a lot more of our work will rise up the abstractions into oversight and understanding. - Luckily, LLMs can help here too because as intelligence and code are increasingly abundant, you can ask for large, custom, discardable software artifacts (e.g. web apps, video explainers) that would have never made sense to create before. Push the boundaries here and you'll be surprised.

  2. Hacker News:AI 热帖54

    Merlin 推出面向小型 AI 原生团队的协作智能体

    Merlin 发布面向小型 AI 原生团队的 AI 智能体,可在 Slack、Linear 和 GitHub 中接收任务并在线程内反馈进展与结果。它通过桌面应用或 CLI 连接团队自有电脑,复用 Claude Code、Codex、现有 MCP servers 和 skills,无需 API keys 或单独模型账单。团队可共享机器、技能、记忆和定时任务,并设置访问控制;5 人以内免费使用。

  3. Hacker News:AI 热帖45

    不要信任,持续验证你的 AI 智能体

    一篇安全草案主张不要信任 AI 智能体,而应将其部署在独立 RasPi 或 mini-PC 上,仅提供明确允许访问的数据。方案通过 Squid SSL Bump MitM 代理、Unbound DoH、auditd、seccomp 和 Landlock 限制并记录网络、DNS 与文件访问,以缩小泄露范围并保留取证审计轨迹。

  4. Hacker News:AI 热帖55

    Skins.dev 发布网页功能定制工具,用 AI 智能体 bryo 为现有 Web 应用添加缺失功能

    Skins.dev 发布浏览器工具,用户可向内置 AI 智能体 bryo 展示并描述希望修改的网页,由其生成只在本地浏览器生效的功能层。该工具无需集成或网站方批准,可添加按钮、列、总计、并排邮件、数据统计等功能,支持 Chrome、Edge、Brave 和 Arc 等 Chromium 浏览器。

  5. Hacker News:AI 热帖57

    Apple 将收紧 macOS 完全磁盘访问控制,应对 AI 智能体带来的隐私风险

    Apple 表示将为 macOS 引入额外隐私控制,让用户只有通过非常明确的操作,才能向应用授予完全磁盘访问权限。Apple 称该权限原本主要为备份应用设计,但部分开发者用它访问文件、邮件、信息和浏览历史,可能在用户未充分理解时暴露系统数据。公告未说明具体控制形式或上线时间,背景是 Meta Muse、OpenAI Dots 等常驻 AI 智能体请求大量个人数据访问。

10月4日周日
  1. Hacker News:AI 热帖47

    如何用 AgentMail AgentID 为 AI 智能体赋予独立身份

    AgentMail 指南提出用 AgentID 替代借用用户登录,为 AI 智能体提供独立邮箱地址、登录密钥,并将用户登记为应用可见的所有者。接入通过 `agentmail inboxes create` 与 `agentmail apps connect` 两条命令完成,支持无头浏览器;登录密钥为 P-256 密钥对,激活后有效期 30 天,删除该密钥即可切断智能体访问。

  2. Elon Musk34

    试试 Grok @Bot

    引用Carnage@0xCarnagee

    this is unreal… the SpaceXAI engineer who leads Grok Bot, Lauren Tan - shipped 2,500 PRs last month - and reviews them after they land, in the morning I took SpaceXAI's Grok Bot docs + her 1-hour livestream and turned them into a 3-page prompting blueprint: the 10-field prompt structure, the 4-rung trust ladder, the two loops, and the guardrails that let agents ship while you sleep. send this PDF to your Grok Bot, save it as a skill, and turn it into an always-on team that actually finishes work

  3. Hacker News:AI 热帖38

    Show HN:Instapath 推出互联网首个个人 AI 智能体竞赛

    Instapath 推出 Personal Agent Challenge,让个人 AI 智能体在线竞逐以比特币锁定的奖金钱包。每个挑战钱包预置 $50 至 $100,首个解开答案并转走资金的智能体获胜,无人认领则滚入下一轮。参与免费,智能体自行创建钱包、监测挑战,赛后将公布各智能体的表现与速度。

  4. Hacker News:AI 热帖37

    资本主义制度下反对 AI 的立场

    一名软件工程师自述从自由派转向社会主义者,并基于社会主义立场反对在资本主义下使用 LLM 与 AI。他承认自己工作中每月花费数百美元使用智能体编程模型,但认为 AI 数据中心加剧气候问题,且 AI 智能体会被资本用来替代劳动者。

  5. elvis28

    AI智能体改变了开发者长期依赖终端的工作模式,用户可将大量操作任务委托给智能体完成,如今也出现了更适合智能体协作的交互界面。终端使用经验培养的系统思维,仍可用于设计沙箱与上层抽象,成为AI时代的能力优势。

    引用Yuchen Jin@Yuchenj_UW

    I was a terminal person for 15+ years. I loved Vim, knew all the shortcuts, and a black terminal made me feel like I was a cool hacker in The Matrix. But terminals assume humans operate computers directly via files, commands, processes. Agents changed all that. You just say your intent. The agent operates the machine, using the same programming languages and Linux commands we spent years learning. I’m still glad I learned computer systems before the AI era though. Understanding what sits beneath the abstraction makes you a better systems thinker. That’s a moat in the AI era.

  6. Hacker News:AI 热帖67

    理解 AI 智能体漂移:模型之外的五类行为变化

    作者指出 AI 智能体即使没有代码、提示词或模型发布,也会因模型、harness、外部系统、自学习和人类使用方式变化而产生行为漂移。文章将漂移分为模型变化、harness 变化、系统变化、智能体学习和人类行为变化五类,并强调错误率不是固定属性。治理上应建立基线、记录完整执行链路,并按变更时、固定周期和线上行为三种节奏持续评估,阈值触发时降低自主性、增加人工审查或停止智能体。

  7. Hacker News 热门(buzzing.cc 中文翻译)52

    pi pod 发布:在自有服务器的隔离沙箱中运行 pi 编码代理

    pi pod 发布为一个围绕 pi 构建的开源项目,可在用户自有服务器的隔离沙箱 pod 中运行 pi 编码代理,并提供可组合环境。项目强调自托管、隐私和可定制,面向希望把智能体工程环境放在自有数据中心的个人与企业;托管服务即将推出,代码位于 github.com/pi-pod/pipod。