2026-06-26（内容获取于 06/26 20:51）

[AINews] OpenAI reports median internal Codex output tokens grew 56x in Research, 32x in Customer Support, 27x in Engineering, and 13x in Legal since November 2025.

Latent Space · 06/26 09:12

It's happening.

中文介绍 OpenAI报告称，自2025年11月以来，其内部Codex模型输出的token数量显著增长。其中，研发部门增长56倍，客户支持部门增长32倍，工程部门增长27倍，法律部门增长13倍，显示出AI工具在各业务领域的深度应用。

Run a vLLM Server on HF Jobs in One Command

Hugging Face Blog · 06/26 08:00

中文介绍 Hugging Face发布新功能，用户现在只需一个命令，即可在Hugging Face Jobs上运行vLLM服务器。此举简化了大型语言模型（LLM）的部署流程，提高了开发效率，方便开发者快速进行模型推理与服务。

The White House is asking OpenAI to slow roll the release of its new model over safety concerns

TechCrunch · 06/26 07:34

OpenAI reportedly plans to share its newest model, GPT 5.6, with a select group of partners instead of with the broader public. The reason: the Trump administration told it to.

中文介绍 据报道，白宫要求OpenAI推迟其最新模型GPT 5.6的公开发布，理由是出于安全担忧。OpenAI计划先将该模型提供给特定的合作伙伴使用，以回应特朗普政府的要求。

YouTube Shorts are getting even shorter with an update that lets you double the playback speed

TechCrunch · 06/26 06:02

YouTube Shorts is getting a makeover.

中文介绍 YouTube Shorts迎来更新，将允许用户以双倍速播放视频。此功能旨在进一步缩短短视频观看时长，提升用户浏览效率和体验，顺应快节奏内容消费趋势。

Patronus AI lands $50M to build ‘digital worlds’ that stress-test AI agents

TechCrunch · 06/26 04:19

Agent-testing startup Patronus AI, founded by former Meta AI researchers, is experiencing nearly insatiable demand, its investor says.

中文介绍 由前Meta AI研究员创立的初创公司Patronus AI获得5000万美元融资，将用于构建“数字世界”以对AI智能体进行压力测试。据投资者称，市场对此类智能体测试服务需求旺盛。

Polymarket says hackers stole users’ funds

TechCrunch · 06/26 03:58

The prediction market giant Polymarket said it's refunding users who had funds stolen due to a third-party breach.

中文介绍 预测市场巨头Polymarket宣布，因第三方系统遭到入侵，导致用户资金被盗。该公司表示，将对受影响的用户进行全额退款，以弥补此次安全漏洞造成的损失。

Xbox follows Apple with price increases

TechCrunch · 06/26 03:39

The company says the increases are being driven by rising memory and console storage prices, with costs more than 2.5x higher than previous levels.

中文介绍 Xbox宣布将提高部分产品价格，公司解释称，此举是由于内存和主机存储成本上涨所致，目前成本已比以往水平高出2.5倍以上。这一涨价行为紧随苹果之后。

Notion Mail shuts down amid agent takeover

TechCrunch · 06/26 03:14

The company said it is discontinuing its email inbox in favor of its AI agent offering as users are increasingly handing over the reins of their email to the agents.

中文介绍 Notion宣布关闭其邮件收件箱服务Notion Mail。公司表示，此举是为了专注于其AI智能体产品，因为用户正越来越多地将电子邮件管理工作交由AI智能体处理，预示着AI在个人效率工具领域的主导地位增强。

a16z-backed Base Power is offering cheaper electricity to the power grid that needs it most

TechCrunch · 06/26 01:52

Base Power is skipping the PJM's troubled interconnection queue by placing its batteries at people's homes, offering backup services in exchange.

中文介绍 由a16z投资的Base Power公司正通过将电池安装在用户家中，为电网提供更廉价的电力。此举旨在解决电力需求痛点，并通过提供备用服务，绕开PJM繁琐的电网互联队列。

Anthropic’s Claude is winning over paid consumers, a market owned by ChatGPT

TechCrunch · 06/26 01:38

Despite ChatGPT's commanding market lead, consumers who pay for AI have been increasingly choosing Anthropic's Claude, data shows.

中文介绍 数据显示，尽管ChatGPT在AI市场占据主导地位，但付费用户正越来越多地选择Anthropic的Claude模型。这表明Claude在特定用户群体中获得了认可，开始在付费AI消费市场中挑战ChatGPT的领先地位。

General Intuition’s $2.3B bet that video games can train AI agents for the real world

TechCrunch · 06/26 00:55

General Intuition has raised $320 million to scale AI trained on millions of hours of gameplay, betting action data can help AI develop something closer to human intuition.

中文介绍 General Intuition公司已筹集3.2亿美元资金，旨在扩大通过数百万小时游戏玩法训练AI智能体的规模。公司押注游戏动作数据能帮助AI智能体发展出更接近人类的直觉，以应对现实世界挑战。

Databricks’ former AI chief thinks he can cut AI’s power bill by 1,000x

TechCrunch · 06/26 00:48

Un-0 is an image-generation system tool that shows for the first time how the company's technology can replicate conventional AI systems.

中文介绍 Databricks前AI负责人提出，其公司技术有望将AI功耗降低1000倍。一款名为Un-0的图像生成系统工具首次展示了该技术如何复制传统AI系统，预示着未来高效能AI的发展方向。

Hacked Klue says criminals are deleting stolen customer data, but now other hackers are making threats

TechCrunch · 06/26 00:40

Market research company Klue told customers that it believes the hacking group that stole their data is now deleting it. The company, however, warned about a second group of hackers wanting ransom.

中文介绍 市场研究公司Klue表示，窃取其客户数据的黑客组织正在删除这些数据。然而，该公司同时警告称，有第二个黑客组织正在威胁索要赎金，使得数据安全形势复杂化。

Which tokens does a hybrid model predict better?

Hugging Face Blog · 06/26 00:11

中文介绍 Hugging Face博客发布文章，探讨混合模型在预测不同类型token时的表现。该研究深入分析了混合模型在特定token上的预测优势，为优化模型性能和理解其工作机制提供了见解。

Rippling now wants to be your entire data stack

TechCrunch · 06/26 00:00

"There were employees doing things like, 'Claude is so helpful for me — it analyzes my calendar and my email and puts together a plan for me,'" he says. "That person was spending at a run rate of $30,000 a year for this."

中文介绍 Rippling公司正寻求成为企业的完整数据栈，提供综合性解决方案。公司创始人指出，有员工每年花费高达3万美元使用AI工具（如Claude）进行日程和邮件分析，这凸显了企业对高效AI成本管理和集成数据平台的需求。

v2.1.193

06/26 05:45

What's changed Added autoMode.classifyAllShell setting to route all Bash/PowerShell commands through the auto-mode classifier instead of only arbitrary-code-execution patterns Added auto-mode denial reasons to the transcript, the denial toast, and /permissions recent denials Added claude_code.assist

中文介绍 Anthropic Claude的代码项目发布v2.1.193版本。主要更新包括新增autoMode.classifyAllShell设置，用于通过自动模式分类器处理所有Bash/PowerShell命令，并在记录和拒绝历史中增加了自动模式拒绝原因。

v2.1.191

06/25 05:58

What's changed Added /rewind support for resuming a conversation from before /clear was run Fixed scroll position jumping to the bottom while reading earlier output during a streaming response Fixed background agents resurrecting after being stopped — stopping an agent from the tasks panel is now pe

中文介绍 Anthropic Claude的代码项目发布v2.1.191版本。更新新增“/rewind”功能，可从“/clear”命令前恢复会话；同时修复了流式响应期间滚动位置跳动和后台代理停止后重新启动等问题。

v2.1.190

06/24 23:53

What's changed Bug fixes and reliability improvements

v2.1.187

06/24 05:03

What's changed Added sandbox.credentials setting to block sandboxed commands from reading credential files and secret environment variables Added org-configured model restrictions to the model picker, --model, /model, and ANTHROPIC_MODEL, with a "restricted by your organization's settings" message w

v2.1.186

06/23 04:37

What's changed Added claude mcp login and claude mcp logout to authenticate MCP servers from the CLI without opening the interactive /mcp menu, with --no-browser stdin redirect support for completing over SSH Added status filtering (press f) to the /workflows agent detail view Added a "Skills" secti

v2.1.185

06/21 04:59

What's changed The stream-stall hint now reads "Waiting for API response · will retry in …" instead of "No response from API · Retrying in …", and triggers after 20s of silence instead of 10s

v2.1.183

06/19 09:20

What's changed Improved auto mode safety: destructive git commands (git reset --hard, git checkout -- ., git clean -fd, git stash drop) are now blocked when you didn't ask to discard local work, git commit --amend is blocked when the commit wasn't made by the agent this session, and terraform destro

v2.1.181

06/18 06:07

What's changed Added /config key=value syntax to set any setting from the prompt (e.g. /config thinking=false) — works in interactive, -p, and Remote Control Added sandbox.allowAppleEvents opt-in setting that lets sandboxed commands send Apple Events on macOS Added CLAUDE_CLIENT_PRESENCE_FILE enviro

v2.1.179

06/17 04:22

What's changed Fixed mid-stream connection drops: partial responses are now preserved instead of showing a raw error, and the spinner no longer gets stuck at "running tool" Fixed mouse-wheel scrolling in WSL2 under Windows Terminal and VS Code (regression in 2.1.172) Fixed a sandbox denyRead/allowRe

v2.1.178

06/16 05:35

What's changed Added Tool(param:value) syntax for permission rules to match a tool's input parameters (with * wildcard), e.g. Agent(model:opus) to block Opus subagents Skills in nested .claude/skills directories now load when working on files there; on a name clash, the nested skill appears as : so

simplex-chat/simplex-chat

Haskell · ★ 11,823 · 🍴 678 · 📈 191 stars today

SimpleX - the first messaging network operating without user identifiers of any kind - 100% private by design! iOS, Android and desktop apps 📱!

中文介绍 SimpleX Chat 是一个开创性的消息网络，其核心设计理念是不使用任何用户标识符，从根本上确保100%的用户隐私。它解决了传统聊天应用中用户身份可追踪的问题，提供极致的匿名通讯体验。项目支持iOS、Android和桌面客户端，适用于高度重视个人隐私的即时通讯用户及对数据安全有严格要求的组织。

google-labs-code/design.md

TypeScript · ★ 20,486 · 🍴 1,690 · 📈 1,475 stars today

A format specification for describing a visual identity to coding agents. DESIGN.md gives agents a persistent, structured understanding of a design system.

中文介绍 DESIGN.md 定义了一种格式规范，旨在帮助编码代理程序理解和应用视觉设计系统。通过提供持久化、结构化的设计描述，它解决了AI在生成用户界面时难以保持设计一致性的挑战。该规范使得AI代理能够更好地遵循品牌指南和设计原则，特别适用于利用AI进行UI/UX开发和自动化设计流程的团队。

commaai/openpilot

Python · ★ 61,635 · 🍴 11,037 · 📈 67 stars today

openpilot is an operating system for robotics. Currently, it upgrades the driver assistance system on 300+ supported cars.

中文介绍 openpilot 是一款面向机器人的开源操作系统，当前主要用于升级300多款受支持汽车的驾驶辅助系统（ADAS）。它能将车辆的原厂辅助驾驶功能提升至更高级别，提供更智能的巡航控制和车道保持能力。该项目为汽车智能化和自动驾驶技术开发提供了灵活的软件平台，适合汽车爱好者、开发者和自动驾驶研究人员。

kunchenguid/no-mistakes

Go · ★ 3,152 · 🍴 187 · 📈 996 stars today

git push no-mistakes

中文介绍 `no-mistakes` 项目很可能是一个用于Git工作流的工具或脚本，通过在`git push`操作前执行一系列检查，帮助开发者避免常见的提交错误。这可能包括代码格式化、静态分析、测试运行或分支校验等，旨在提高代码质量和团队协作效率。它特别适合需要强制执行代码提交规范的开发团队和追求无误提交的个人开发者。

grafana/grafana

TypeScript · ★ 74,724 · 🍴 14,116 · 📈 17 stars today

The open and composable observability and data visualization platform. Visualize metrics, logs, and traces from multiple sources like Prometheus, Loki, Elasticsearch, InfluxDB, Postgres and many more.

中文介绍 Grafana 是一款开源且可组合的可观测性与数据可视化平台。它能够聚合并展示来自Prometheus、Loki、Elasticsearch、InfluxDB、Postgres等多种数据源的指标、日志和追踪数据。通过直观的仪表盘，Grafana帮助用户监控系统性能、快速定位故障并进行业务数据分析，是DevOps工程师、SRE团队和系统管理员进行运维监控与数据洞察的核心工具。

ripienaar/free-for-dev

HTML · ★ 123,515 · 🍴 13,039 · 📈 48 stars today

A list of SaaS, PaaS and IaaS offerings that have free tiers of interest to devops and infradev

中文介绍 free-for-dev 维护了一份精心策划的列表，收录了众多为开发者和运维工程师提供免费使用额度（Free Tier）的SaaS、PaaS和IaaS服务。它旨在帮助开发人员和基础设施工程师快速发现并利用免费资源，从而有效降低开发成本、加速项目启动。这份列表是个人开发者、初创团队和学生寻找免费服务的重要参考资料。

opendatalab/MinerU

Python · ★ 70,078 · 🍴 5,922 · 📈 644 stars today

Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.

中文介绍 MinerU 致力于将PDF、Office文档等复杂非结构化文件转换为LLM（大型语言模型）友好的Markdown或JSON格式，以支持Agentic工作流。它解决了LLM直接处理复杂文档的难题，通过高效的数据提取和结构化，赋能AI代理进行更精准的信息理解和任务执行，特别适用于构建智能文档分析和自动化系统的AI开发者与数据科学家。

alchaincyf/zhangxuefeng-skill

★ 9,158 · 🍴 2,554 · 📈 220 stars today

张雪峰.skill — 张雪峰的认知操作系统。高考志愿/考研/职业规划的实战思维框架。由女娲.skill生成。

中文介绍 张雪峰.skill 是一个基于知名教育咨询师张雪峰理念的“认知操作系统”，旨在为高考志愿、考研及职业规划提供一套实战性的思维框架。该项目由“女娲.skill”生成，可能是一个基于AI代理的知识应用，帮助用户系统性地分析和决策人生重要选择，特别适合面临升学和职业发展困惑的学生及个人寻求智能辅助。

mauriceboe/TREK

TypeScript · ★ 7,249 · 🍴 634 · 📈 241 stars today

A self-hosted travel/trip planner with real-time collaboration, interactive maps, PWA support, SSO, budgets, packing lists, and more.

中文介绍 TREK 是一款可自行托管的旅行计划工具，提供实时协作、交互式地图、PWA支持、单点登录、预算管理和行李清单等丰富功能。它旨在解决旅行规划中信息分散和团队协作不便的问题，同时允许用户完全掌控数据隐私。TREK适合重视隐私、寻求自助托管解决方案，并需要与朋友或家人高效协作规划旅行的个人和团体。

xbtlin/ai-berkshire

Python · ★ 2,817 · 🍴 411 · 📈 309 stars today

AI 时代的伯克希尔：基于 Claude Code 的价值投资研究框架。巴菲特·芒格·段永平·李录四大师方法论 + 多Agent并行研究。| AI-era Berkshire: a value investing research framework built on Claude Code. 4 masters' methodologies + multi-agent adversarial analysis.

中文介绍 AI 时代的伯克希尔是一个基于 Claude Code 构建的价值投资研究框架，融合了巴菲特、芒格、段永平、李录四位投资大师的方法论，并通过多Agent并行研究提升投资分析的效率与深度。该项目旨在利用AI技术赋能价值投资，为投资者提供系统化的研究工具，适用于寻求AI辅助决策的专业投资者、金融分析师及相关研究人员。

calesthio/OpenMontage

Python · ★ 23,031 · 🍴 2,562 · 📈 3,434 stars today

World's first open-source, agentic video production system. 12 pipelines, 52 tools, 500+ agent skills. Turn your AI coding assistant into a full video production studio.

中文介绍 OpenMontage 是全球首个开源的Agentic视频制作系统，它能将AI编程助手转变为一个功能完备的视频制作工作室。该系统集成了12条生产线、52个工具和500多个Agent技能，旨在自动化和简化从构思到最终发布的视频创作全流程。它赋能内容创作者和AI开发者高效生成高质量视频内容，大幅提升视频制作效率。

aws/agent-toolkit-for-aws

Python · ★ 1,257 · 🍴 117 · 📈 47 stars today

Official, AWS-supported MCP servers, skills, and plugins to help AI agents build on AWS

中文介绍 agent-toolkit-for-aws 是由AWS官方支持的AI代理工具包，旨在帮助开发者在AWS云上更高效地构建和部署AI代理应用。它提供了MCP服务器、预定义技能和插件等资源，简化了AI代理与AWS服务的集成过程。该工具包使开发者能够利用AWS强大的基础设施，快速开发智能、可扩展的AI代理解决方案，适用于构建企业级AI应用。

NanmiCoder/MediaCrawler

Python · ★ 53,222 · 🍴 10,956 · 📈 398 stars today

小红书笔记 | 评论爬虫、抖音视频 | 评论爬虫、快手视频 | 评论爬虫、B 站视频 ｜ 评论爬虫、微博帖子 ｜ 评论爬虫、百度贴吧帖子 ｜ 百度贴吧评论回复爬虫 | 知乎问答文章｜评论爬虫

中文介绍 MediaCrawler 提供了一系列针对中国主流社交媒体和视频平台的爬虫工具，涵盖小红书、抖音、快手、B站、微博、百度贴吧和知乎的帖子、视频及评论数据。它解决了从这些平台批量、高效地获取公开数据的需求，为数据分析、舆情监控和市场研究提供了便利，适用于数据科学家、市场分析师和研究人员进行社交媒体数据采集。

garrytan/gstack

TypeScript · ★ 116,271 · 🍴 17,226 · 📈 767 stars today

Use Garry Tan's exact Claude Code setup: 23 opinionated tools that serve as CEO, Designer, Eng Manager, Release Manager, Doc Engineer, and QA

中文介绍 gstack 是风投家 Garry Tan 分享的 Claude Code 配置，包含23个“有主见的”AI工具，能分别扮演CEO、设计师、工程经理、发布经理、文档工程师及QA等多种角色。它提供了一套利用AI代理实现复杂业务流程自动化的框架，旨在帮助用户高效地搭建和管理AI驱动的工作流，特别适合希望借助AI提升团队效率的企业家和开发者。

IceWhaleTech/CasaOS

Go · ★ 35,105 · 🍴 2,008 · 📈 223 stars today

CasaOS - A simple, easy-to-use, elegant open-source Personal Cloud system.

中文介绍 CasaOS 是一款简洁、易用且优雅的开源个人云系统，旨在帮助用户轻松搭建和管理自己的家庭服务器或个人存储。它解决了传统NAS系统配置复杂的问题，提供友好的图形界面和应用商店，让用户能够便捷地存储文件、运行各种服务，是个人数据管理、家庭媒体中心和自托管应用的理想解决方案，特别适合普通家庭用户。

JCodesMore/ai-website-cloner-template

TypeScript · ★ 21,015 · 🍴 3,061 · 📈 1,024 stars today

Clone any website with one command using AI coding agents

中文介绍 ai-website-cloner-template 提供了一个通过AI编码代理一键克隆任何网站的解决方案。它旨在显著简化网站的复刻过程，将复杂的网页结构、样式和内容转换为可编辑的代码，极大提升了前端开发和原型设计的效率。该工具特别适用于需要快速构建类似网站、进行设计复现或学习他人网站结构的前端开发者和AI应用研究者。

Panniantong/Agent-Reach

Python · ★ 41,872 · 🍴 3,325 · 📈 1,547 stars today

Give your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.

中文介绍 Agent-Reach 旨在赋予AI代理“看见”整个互联网的能力，通过一个命令行接口（CLI），让AI代理能免费访问并搜索Twitter、Reddit、YouTube、GitHub、Bilibili、小红书等平台内容。它解决了AI代理获取实时网络信息受限或需支付高额API费用的痛点，极大地扩展了AI代理的信息获取范围和应用场景，适合AI开发者和研究人员构建更智能的Agent。

Loops explained: Claude, GPT, Mira and what actually works

@AnatoliKopadze · 83.0K 粉丝 · 1.3M 阅 · 584 赞 · 70 转

06/20 21:40

AI has been in everyone's hands for years. Most people who use it every day still use it the slowest way there is: type a request, wait, fix it, ask again, all by hand. Not because the faster way is

中文介绍 该帖解释了 AI 循环（loops）的概念，指出当前多数人使用 AI 效率低下的问题，并对比 Claude、GPT、Mira 等模型在循环工作流中的实际效果和应用。强调通过自动化流程提升 AI 使用效率，避免手动重复操作。

How to Build an AI Second Brain With Claude and Obsidian That Gets Smarter Every Day (Full Guide)

@undefinedKi · 3.9K 粉丝 · 1.0M 阅 · 601 赞 · 78 转

06/20 20:15

Your best ideas are scattered across a dozen places right now. Notes apps. Browser tabs. Old chats with Claude that you closed and will never find again. Every time you sit down to work, you rebuild

中文介绍 一份完整指南，教用户如何利用 Claude 和 Obsidian 构建一个「AI 第二大脑」。该系统能整合分散的笔记和想法，通过 AI 辅助整理和学习，实现每天「变聪明」，从而解决信息管理和知识积累的痛点。

40 Claude Opus 4.8 Workflows That Make Money While You Sleep

@eng_khairallah1 · 69.3K 粉丝 · 678.7K 阅 · 506 赞 · 81 转

06/21 17:00

Most people are using Claude Opus 4.8 to answer questions. Save this :) A small group of people have it running businesses while they sleep. The difference is not the model. You both have access to

中文介绍 博主分享了 40 种利用 Claude Opus 4.8 构建自动化工作流的方法，旨在帮助用户在「睡觉时也能赚钱」。这些工作流超越了简单的问答模式，通过深度利用 AI 代理能力实现商业运营和收入生成。

The efficiency gap: How do cells and GPUs compare when running the exact same algorithm?

@spandan_madan · 1.1K 粉丝 · 626.9K 阅 · 516 赞 · 28 转

06/24 02:13

AI hardware is having a moment. Hyperscaler capex on AI data centres is on track to clear $690 billion in 2026, and private equity has followed in scale — Blackstone alone reports a $55B+ data-centre

中文介绍 该帖探讨 AI 硬件领域热点，比较生物细胞与 GPU 在执行相同算法时的效率差距。文章揭示 AI 数据中心巨大的投资规模（2026 年超 6900 亿美元），并指出当前硬件发展面临的挑战与机遇。

How To Learn Anything 10x Faster Using Claude

@sairahul1 · 120.1K 粉丝 · 581.3K 阅 · 516 赞 · 110 转

06/20 16:30

Learning anything today is easy and confusing at the same time. Easy because AI can explain almost anything in seconds. Confusing because most people just ask random questions, get random answers, and

中文介绍 该帖分享利用 Claude 提升学习效率的策略，旨在帮助用户将学习速度提高 10 倍。文章指出，尽管 AI 能快速解答问题，但若缺乏系统性提问，效果甚微。博主提供了更有效的 AI 辅助学习方法。

Introducing Engram: Scaling compute on your context

@EngramLab · 1.2K 粉丝 · 255.7K 阅 · 537 赞 · 76 转

06/24 01:01

We’re Engram. We’re building AI that learns from you and deeply understands your work. Today’s AI models don’t understand what you do. Not really. Everything models know comes from their training –

中文介绍 Engram 团队发布新产品 Engram，旨在构建一款能够「深度理解用户工作」并「从用户那里学习」的 AI。它解决了现有 AI 模型无法真正理解用户个性化上下文的痛点，通过上下文计算扩展提升 AI 的实用性。

30 Core Agentic Engineering Concepts Every Developer Should Know

@sairahul1 · 120.1K 粉丝 · 249.0K 阅 · 502 赞 · 90 转

06/23 17:28

There are 8 billion people on the planet. Only a fraction of developers understand how AI agents actually work. Not the demos. Not the hype. The real engineering underneath. Every week a new agent

中文介绍 博主罗列了 30 个 AI 代理工程的核心概念，旨在帮助开发者深入理解 AI 代理的实际工作原理。文章强调，这些概念超越了表面演示和炒作，揭示了 AI 代理背后的工程实践，对于掌握其技术基础至关重要。

ORACLE: Official AI Agents Trade on Polymarket

@Oracle_Trade_ai · 39.9K 粉丝 · 197.8K 阅 · 2.8K 赞 · 580 转

06/22 16:52

In 2026, autonomous AI agents have become one of the most effective strategies on prediction markets. Over 30% of all activity on Polymarket now comes from algorithmic and AI-powered wallets. We

中文介绍 ORACLE 宣布推出官方 AI 代理，专注于在预测市场 Polymarket 上进行交易。帖子展望 2026 年，届时自主 AI 代理将成为预测市场最有效的策略之一，其交易量已占 Polymarket 总活跃度的 30% 以上。

Why we're bullish on loops

@posthog · 21.8K 粉丝 · 162.4K 阅 · 512 赞 · 36 转

06/24 01:26

When the creators of both OpenClaw and Claude Code speak, people listen. And recently Peter Steinberger and Boris Cherny have both been talking about the same concept: loops. Their argument? You

中文介绍 Posthog 表达了对 AI「循环」（loops）概念的看好，引用 OpenClaw 和 Claude Code 创始人对该概念的认可。文章指出，AI 循环代表了提升 AI 效率和自动化能力的关键，是未来 AI 应用的重要方向。

ORACLE: Official AI Agents Trade on Polymarket

@OracAItrading · 31.8K 粉丝 · 141.6K 阅 · 2.8K 赞 · 576 转

06/22 16:52

In 2026, autonomous AI agents have become one of the most effective strategies on prediction markets. Over 30% of all activity on Polymarket now comes from algorithmic and AI-powered wallets. We

中文介绍 ORACLE 宣布推出官方 AI 代理，专注于在预测市场 Polymarket 上进行交易。帖子展望 2026 年，届时自主 AI 代理将成为预测市场最有效的策略之一，其交易量已占 Polymarket 总活跃度的 30% 以上。

Interactions API: Our primary interface for Gemini models and agents

@GoogleAIStudio · 179.4K 粉丝 · 138.2K 阅 · 504 赞 · 42 转

06/23 01:21

Today we're announcing that the Interactions API has reached general availability and is now our primary API for interacting with Gemini models and agents. We launched its public beta in December

中文介绍 GoogleAIStudio 宣布 Interactions API 已正式发布并达到通用可用性（GA），现已成为与 Gemini 模型和 AI 代理交互的主要接口。该 API 自去年 12 月公开测试以来，为开发者提供了更稳定、便捷的 Gemini 模型集成方案。

ORACLE: Official AI Agents Trade on Polymarket

@Oractrading · 33.9K 粉丝 · 109.2K 阅 · 2.8K 赞 · 585 转

06/22 16:20

In 2026, autonomous AI agents have become one of the most effective strategies on prediction markets. Over 30% of all activity on Polymarket now comes from algorithmic and AI-powered wallets. We

中文介绍 ORACLE 宣布推出官方 AI 代理，专注于在预测市场 Polymarket 上进行交易。帖子展望 2026 年，届时自主 AI 代理将成为预测市场最有效的策略之一，其交易量已占 Polymarket 总活跃度的 30% 以上。

How To Use Loop Engineering To Build A Self-Improving Quant Trading System

@RohOnChain · 51.4K 粉丝 · 108.3K 阅 · 501 赞 · 65 转

06/22 21:54

I will break down exactly how to build the loops that run an entire quant trading system on their own. Let's get straight to it. Bookmark This - I'm Roan, a backend developer working on system

中文介绍 博主详细讲解如何运用「循环工程」（Loop Engineering）来构建一个「自我改进的量化交易系统」。该系统能够自主运行，通过循环机制不断优化交易策略，旨在实现自动化和高效的量化交易。

WTF Is a Loop? Part 2: The 15 Loops People Are Actually Running (and the Commands to Steal Them)

@mvanhorn · 35.2K 粉丝 · 102.4K 阅 · 510 赞 · 56 转

06/21 04:09

Earlier this month I wrote WTF Is a Loop? Peter Steinberger vs. Boris Cherny, which did 3.6M views on what a loop even is. This is the sequel, and it answers the next question: which loops do people

中文介绍 继上月关于 AI「循环」概念的病毒式帖子后，博主发布续篇，揭示了目前用户实际运行的 15 种 AI 循环，并提供了直接可用的命令。文章旨在普及 AI 循环的实用应用，帮助读者理解并掌握这些高效工作流。

From Prompting Agents to Loop Engineering

@omarsar0 · 308.0K 粉丝 · 90.2K 阅 · 504 赞 · 69 转

06/20 00:31

A claim has been circulating in AI coding circles: stop prompting your coding agents and start designing loops that prompt them for you. As with everything new, this stuff gets repeated often and

中文介绍 该帖探讨 AI 编码领域从「提示词代理」（Prompting Agents）到「循环工程」（Loop Engineering）的转变。文章提出，与其手动向 AI 代理发送提示，不如设计自动化的循环来生成提示，从而提高编码效率和自动化水平。

How GLM-5.2 Beat Fable 5 at Website Design

@Designarena · 13.9K 粉丝 · 80.4K 阅 · 518 赞 · 39 转

06/20 01:58

GLM 5.2 ranks 1st overall on Design Arena’s single-turn, HTML Web Design (Non-Agentic) evaluation, 5 places higher than its predecessor GLM-5.1. To do so, it beat Claude Fable 5, Opus 4.6, and Opus

中文介绍 Designarena 发布评估结果，显示 GLM-5.2 在单轮 HTML 网页设计（非代理）评估中总体排名第一，超越了 Claude Fable 5、Opus 4.6 及 Opus。GLM-5.2 比其前身 GLM-5.1 提升了 5 个名次，展现卓越的网页设计能力。

Subnets mine TAO; TAO mines Subnets

@const_reborn · 29.7K 粉丝 · 79.8K 阅 · 503 赞 · 116 转

06/23 09:21

e_i \;\propto\; \underbrace{\rho_i \times \bar{p}_i}_{\text{linear (maximize)}} \times \underbrace{(1 - b_i)}_{\text{boolean gate}} Disclaimer: this upgrade only effects subnet owners and dynamic TAO

中文介绍 该帖围绕 Bittensor 网络中的「Subnets 挖 TAO，TAO 挖 Subnets」经济激励机制展开讨论。博主提出了一个涉及数学模型的 Subnet 收益计算方式，并强调此次升级主要影响 Subnet 持有者和动态 TAO 机制。

How we built the world’s fastest API for GLM-5.2

@philipkiely · 8.6K 粉丝 · 73.1K 阅 · 544 赞 · 36 转

06/23 08:13

GLM-5.2 is the biggest news in open models since DeepSeek-R1. It’s easy to see why. GLM-5.2 delivers comparable performance to GPT 5.5 and Opus 4.8 at a fraction of the cost, generally 70-80% less

中文介绍 博主分享了如何为 GLM-5.2 构建「全球最快 API」的经验。GLM-5.2 因其与 GPT 5.5 和 Opus 4.8 媲美的性能，同时成本降低 70-80%，被誉为 DeepSeek-R1 之后开放模型领域的最大新闻。

The Next 5 Years: A Supersonic Tsunami

@PeterDiamandis · 400.3K 粉丝 · 51.7K 阅 · 551 赞 · 75 转

06/21 23:22

Elon described the near future as a “supersonic tsunami”: a wave moving so fast and so large that by the time you hear it coming, it has already broken over you. The phrase stuck with me. So let me

中文介绍 Peter Diamandis 引用埃隆·马斯克对未来五年的描述：「超音速海啸」，强调科技发展速度之快、影响之巨。文章展望了未来技术浪潮将以惊人速度席卷而来，提醒人们必须适应并把握这些即将到来的变革。

Loop Engineering: Build an AI That Codes While You Sleep

@phosphenq · 12.4K 粉丝 · 49.4K 阅 · 506 赞 · 63 转

06/21 06:59

Last December, Boris Cherny shipped 259 pull requests in a single month. Every one written by Claude. He says he did not open an editor the whole time. His job was not to write code. It was to write

中文介绍 该帖聚焦「循环工程」，展示了如何构建一个能在用户「睡觉时自动编码」的 AI 系统。以 Boris Cherny 为例，他上月通过 Claude 自动生成 259 个代码提交，全程无需手动编辑，强调了循环工程在自动化编程中的巨大潜力。

DanceOPD: On-Policy Generative Field Distillation

👍 50

06/25 08:00

Modern image generation demands a single model that unifies diverse capabilities, including text-to-image (T2I), local editing, and global editing. However, these capabilities are rarely naturally aligned and often conflict. For instance, editing tends to degrade T2I performance, while global and lo

中文介绍 DanceOPD模型旨在解决现代图像生成中文本到图像（T2I）、局部编辑和全局编辑等多样能力间的冲突。这些能力往往难以自然对齐，例如编辑操作可能降低T2I性能。该研究提出一种策略性生成场蒸馏方法，以期在一个统一模型中实现多功能性，并克服现有技术中各能力相互影响的问题。

Ask, Solve, Generate: Self-Evolving Unified Multimodal Understanding and Generation via Self-Consistency Rewards

👍 0

06/25 08:00

Most unified large multimodal models (LMMs) that support both visual understanding and image generation still rely on curated post-training supervision, such as human annotations, preference labels, or external reward models. We ask whether a unified LMM can improve both abilities autonomously using

中文介绍 现有多数统一大语言多模态模型（LMMs）依赖人工标注或外部奖励模型进行后训练监督。本研究提出「问、解、生成」的自进化框架，通过自洽奖励机制，探讨LMMs在纯无监督设置下能否同时提升视觉理解和图像生成能力，以实现模型的自我演进，减少对人工监督的依赖。

Paying More Attention to Visual Tokens in Self-Evolving Large Multimodal Models

👍 0

06/25 08:00

Recently, self-evolving large multimodal models (LMMs) have received attention for improving visual reasoning in a purely unsupervised setting. However, multi-role self-play and self-consistency reward schemes in existing self-evolving LMMs optimize answer agreement without ensuring the decoder atte

中文介绍 近期，自进化大语言多模态模型（LMMs）因在纯无监督环境下提升视觉推理而受关注。但现有自博弈和自洽奖励机制主要优化答案一致性，未充分确保视觉信息的准确性。本研究提出应更关注视觉token，以期在自进化LMMs中提升其视觉推理的准确性，从而实现更可靠的视觉理解。

PhysiFormer: Learning to Simulate Mechanics in World Space

👍 2

06/25 08:00

We present PhysiFormer, a diffusion transformer for physically-plausible 3D object motion. Unlike video world models that operate in view-dependent pixel space, PhysiFormer represents objects as 3D meshes expressed in world coordinates. Given the initial vertex positions and velocities, as well as o

中文介绍 论文提出PhysiFormer，这是一种用于生成物理可信3D物体运动的扩散Transformer模型。与在依赖视角像素空间操作的视频世界模型不同，PhysiFormer将物体表示为世界坐标系中的3D网格。通过输入初始顶点位置和速度，该模型旨在更准确地模拟真实世界的物理运动。

Autoregressive Boltzmann Generators

👍 0

06/25 08:00

Efficient sampling of molecular systems at thermodynamic equilibrium is a hallmark challenge in statistical physics. This challenge has driven the development of Boltzmann Generators (BGs), which allow rapid generation of uncorrelated equilibrium samples by combining a generative model with exact li

中文介绍 在统计物理学中，高效采样热力学平衡态分子系统是一项重要挑战。为解决此问题，玻尔兹曼生成器（BGs）应运而生。本文介绍一种新型自回归玻尔兹曼生成器，它结合生成模型，能快速生成不相关的平衡态样本，旨在显著提升分子系统模拟的效率和准确性。

SAM2Matting: Generalized Image and Video Matting

👍 0

06/25 08:00

Despite impressive advances in image matting, video matting remains challenging due to the inherent gap between high-level tracking, which requires frame-wise understanding, and low-level matting, which focuses on extremely fine-grained details. Existing methods attempt this with expensive and narro

中文介绍 尽管图像抠图技术取得显著进展，但视频抠图仍具挑战，主要源于高层级跟踪（需帧间理解）与低层级抠图（需精细细节）间的固有鸿沟。本文提出SAM2Matting，旨在提供一个通用的图像和视频抠图解决方案，以弥合这一差距，并有效处理不同粒度的视觉信息，实现更广义的抠图能力。

Hallucination in World Models is Predictable and Preventable

👍 1

06/25 08:00

Modern generative world models render increasingly realistic action-controllable futures, yet they frequently hallucinate: rollouts remain visually fluent while drifting from the ground-truth dynamics. We hypothesize that hallucination concentrates in low-coverage regions of the state-action space,

中文介绍 现代生成式世界模型能渲染出越来越真实的、可控的未来场景，但常出现幻觉，即视觉流畅却偏离真实动力学。本研究假设幻觉集中在状态空间的低覆盖区域，并提出幻觉是可预测和可预防的。这为通过改进模型机制减少甚至消除这些不准确的生成提供了新的研究方向。

OctoSense: Self-Supervised Learning for Multimodal Robot Perception

👍 0

06/25 08:00

We present OctoSense, an open-source sensor platform with stereo RGB and event cameras, LiDAR, a thermal camera, an inertial measurement unit, RTK-corrected global positioning system, and proprioception (CAN bus data from a car, and joint angles for a quadruped robot). The eponymous OctoSense datase

中文介绍 论文介绍OctoSense，一个开源多模态机器人感知平台。它集成了双目RGB和事件相机、激光雷达、热像仪、惯性测量单元、RTK纠正全球定位系统，以及本体感知（包括车辆CAN总线数据和四足机器人关节角度）。该平台旨在通过自监督学习，全面提升机器人的多模态感知能力。

ViQ: Text-Aligned Visual Quantized Representations at Any Resolution

👍 34

06/25 08:00

A unified representation for text and vision is a natural pursuit, as it enables simpler multimodal modeling and more efficient training. However, representing images as discrete signals in the same way as text inevitably introduces severe information loss. Existing work struggles to balance low-lev

中文介绍 统一文本与视觉的表征有助于简化多模态建模和提高训练效率。然而，将图像离散化为与文本相同的信号不可避免地会引入严重信息损失。本文提出ViQ模型，旨在克服现有工作在任意分辨率下保持信息完整性的挑战，实现文本与视觉的对齐量化表征，提升多模态模型的效率。

When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models

👍 1

06/25 08:00

Multi-model LLM systems such as routing, voting, cascades, fusion, and mixture-of-agents are used to beat single-model accuracy. We show that their gain is capped by a quantity the field rarely reports. For any policy whose output is one member model answer, accuracy cannot exceed one minus beta, wh

中文介绍 多模型LLM系统如路由、投票和代理混合，常用于超越单模型准确性。本研究通过对67个前沿模型进行分析，揭示了这类系统在提升准确性方面的上限，即共同失败上限。研究指出，当系统输出为单个成员模型的答案时，其准确性无法超过这一上限，挑战了业界现有认知。

How Much Static Structure Do Code Agents Need? A Study of Deterministic Anchoring

👍 0

06/25 08:00

LLM-based code agents navigate repositories through keyword search but miss the structural relationships, such as call graphs, inheritance hierarchies, and configuration dependencies, that define how software actually works. This makes agent navigation stochastic and difficult to reproduce across ru

中文介绍 基于LLM的代码代理目前主要通过关键词搜索浏览代码库，但常忽略调用图、继承层次和配置依赖等软件结构关系，导致导航随机且效率低下。本研究探讨了代码代理需要多少静态结构信息，并提出确定性锚定方法，旨在提升代理对代码库的理解和导航能力，使其更高效地执行任务。

To Run or Not to Run: Analyzing the Cost-Effectiveness of Code Execution in LLM-Based Program Repair

👍 1

06/25 08:00

LLM-based agents for program repair are increasingly built on a "generate-run-revise" paradigm, iteratively executing tests to evaluate and refine patches. This execution-based approach has become standard practice in state-of-the-art systems. However, executions can be time-consuming and expensive,

中文介绍 基于LLM的程序修复代理日益采用「生成-运行-修订」范式，通过迭代执行测试来评估和改进补丁。尽管这种基于执行的方法已成为先进系统的标准实践，但代码执行可能耗时且成本高昂。本研究分析了LLM程序修复中代码执行的成本效益，旨在探讨其在实际应用中的有效性和潜在优化空间。

PhysRAG: Enhancing Physics-Awareness in Video Generation via Retrieval-Augmented Generation

👍 0

06/25 08:00

Developing physically aware video generation models remains a significant challenge due to the difficulty in capturing diverse physical phenomena, such as thermal dynamics, mechanics, and optics. In this work, we introduce PhysRAG, a novel pipeline that enhances physical awareness in video generatio

Qwen-Image-Agent: Bridging the Context Gap in Real-World Image Generation

👍 29

06/25 08:00

While text-to-image (T2I) models have achieved remarkable progress, they struggle with real-world requests that are often underspecified, implicit, or dependent on up-to-date knowledge. We identify this challenge as the Context Gap: the mismatch between the user context and the sufficient generation

Confidence-Aware Tool Orchestration for Robust Video Understanding

👍 6

06/25 08:00

Video reasoning language models implicitly assume that every input frame is equally reliable. This leads to what we term the Blind Trust Problem: under realistic perturbations such as motion blur, glare, or occlusion, frontier video reasoning models can suffer 15-30%p accuracy drops on real-world em

A Pipeline for Generating Longitudinal Synthetic Clinical Notes Using Large Language Models

👍 0

06/25 08:00

Synthetic data is increasingly used to enable the development and evaluation of AI systems in domains where access to real-world data is restricted. In healthcare, clinical documentation presents particular challenges due to its sensitivity. This work introduces a synthetic clinical notes pipeline a

AgentX: Towards Agent-Driven Self-Iteration of Industrial Recommender Systems

👍 0

06/25 08:00

Recommendation algorithm iteration is moving from an artisanal, engineer-bound process toward an industrialized research loop, but this transition remains blocked by a structural execution bottleneck: the idea-to-launch cycle still depends on human engineers to generate hypotheses, modify production

KARLA: Knowledge-base Augmented Retrieval for Language Models

👍 0

06/25 08:00

We propose a new method that allows an LLM to automatically pull in factual knowledge from a knowledge base during token generation. This means that (1)~factual knowledge in the LLM output can be updated without retraining the LLM, (2)~facts in the LLM output can be traced to the knowledge base for

OPID: On-Policy Skill Distillation for Agentic Reinforcement Learning

👍 27

06/25 08:00

Outcome-based reinforcement learning provides a stable optimization backbone for language agents, but its sparse trajectory-level rewards provide little guidance on which intermediate decisions should be reinforced or suppressed. On-policy self-distillation offers dense token-level supervision, yet

DiCoBench: Benchmarking Multi-Image Fine-Grained Perception via Differential and Commonality Visual Cues

👍 0

06/25 08:00

Recent advancements in Multimodal Large Language Models (MLLMs) have demonstrated impressive fine-grained perception capabilities. However, existing benchmarks predominantly rely on explicit textual cues or low-resolution inputs, failing to evaluate a model's ability to autonomously perceive implici

PhyEditBench: A Real-World Multi-Stage Benchmark for Physics-Aware Image Editing

👍 0

06/25 08:00

While instruction-based image editing, enabled by multi-modal generative models, has advanced significantly, existing benchmarks lack a comprehensive evaluation of physics-based reasoning, a critical capability for handling real-world scenarios. To address this, we introduce PhyEditBench, a benchmar

Nemotron-TwoTower: Diffusion Language Modeling with Pretrained Autoregressive Context

👍 0

06/25 08:00

Diffusion language models offer a promising alternative to autoregressive models due to their potential for parallel and iterative generation. However, existing approaches use a single network for both context representation and iterative denoising, forcing one model to serve both roles and limiting

In-Context World Modeling for Robotic Control

👍 5

06/25 08:00

Modern Vision-Language-Action (VLA) models often fail to generalize to novel setups, such as altered camera viewpoints or robot morphologies, because they are typically conditioned only on current observations and language instructions. By ignoring the underlying system configuration as a variable,

JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Parallel Tree Drafting

👍 18

06/25 08:00

Speculative decoding (SD) accelerates autoregressive Large Language Models (LLMs) by drafting multiple tokens and verifying them in parallel, but it faces a scaling limitation: increasing the draft budget improves speed only when acceptance remains high and drafting overhead stays low. This ceiling

Running the Gauntlet: Re-evaluating the Capabilities of Agents Beyond Familiar Environments

👍 9

06/25 08:00

As agentic systems continue to evolve and are widely deployed in real-world scenarios, there is a growing demand to faithfully evaluate their capabilities. However, current benchmarks are typically built on popular applications with relatively simple tasks and focus on a narrow set of capabilities w

AnySimLite: A Lightweight Few-Shot Similarity Encoder for On-Device Speech-Adjacent Classification

👍 0

06/24 08:00

To minimize privacy concerns and inference latency on edge devices like smartphones, lightweight on-device models remain important for end-user applications. Many of these applications involve natural language classification, but deploying multiple specialized models creates a memory footprint chall

Listening Like a Judge: A Music-Aware Framework for Automatic Singing Performance Evaluation

👍 0

06/24 08:00

Automatic singing quality assessment (SQA) requires evaluating lyrical correctness and musical fidelity while handling expressive variations. However, existing systems largely rely on either acoustic cues or lyric transcriptions exclusively, limiting holistic performance evaluation. Furthermore, the

The Verification Horizon: No Silver Bullet for Coding Agent Rewards

👍 20

06/24 08:00

A classical intuition holds that verifying a solution is easier than producing one. For today's coding agents, this intuition is being inverted: as foundation models develop stronger reasoning capabilities and engineering harnesses grow more sophisticated, generating complex candidate solutions is n

COrigami: An AI Pipeline for Co-Designing Flat-Foldable Visually Recognisable Origami

👍 1

06/24 08:00

While generative AI has achieved remarkable success in solving problems with verifiable solutions, generating physical art that satisfies both strict geometric constraints and subjective visual aesthetics remains a challenge. This paper presents an approach to tackle these difficulties in the domain

Fast LeWorldModel

👍 3

06/24 08:00

Joint-Embedding Predictive Architectures (JEPAs), including recent LeWorldModel (LeWM), have become a promising foundation for reconstruction-free visual world models. For visual planning, however, LeWM evaluates candidate action sequences by repeatedly applying a local one-step latent transition mo

Aurora Notch

06/26 13:37

A private notch workspace for every Mac

中文介绍 「Aurora Notch」是一款专为Mac用户设计的私有刘海区工作空间应用。它利用Mac屏幕的刘海区域，提供一个私密的工作环境，帮助用户高效管理和使用空间。

Basedash for Excel

06/26 12:15

Turn any Excel file into a live dashboard

中文介绍 「Basedash for Excel」是一款工具，能将任意Excel文件转换成实时数据仪表板。它旨在帮助用户更直观、动态地展示和分析Excel数据，提升数据可视化效率。

Group Subscriptions by beehiiv

06/26 09:59

Sell subscriptions to teams, companies, and organizations.

中文介绍 「beehiiv」推出的群组订阅功能，允许用户向团队、公司和组织出售订阅服务。这为内容创作者和平台提供了一种新的商业模式，以便更好地服务企业级客户。

Cewsco

06/26 04:33

All-in-one AI assistant — chat, images, voice & market data

中文介绍 「Cewsco」是一款一体化AI助手，集成了聊天、图像生成、语音交互和市场数据分析等多种功能。它旨在为用户提供全面的AI支持，满足不同场景下的需求。

LockIn MCP

06/25 17:35

Let AI block distractions for you when you need to lock in

中文介绍 「LockIn MCP」是一款利用AI技术帮助用户专注工作的工具。当用户需要集中精力时，它能自动屏蔽干扰，提供一个无中断的工作环境，从而提升效率。

Sleek Analytics

06/25 13:52

See who's on your site. Right now.

中文介绍 「Sleek Analytics」提供实时网站访客分析服务，让用户能够立即查看当前网站的活跃访客情况。它旨在帮助网站管理员和营销人员了解用户行为。

Gemini Spark

06/25 13:47

Your 24/7 personal AI agent

中文介绍 「Gemini Spark」是一款24/7全天候个人AI智能体。它作为用户的常驻助手，旨在提供持续的智能服务和支持，协助处理各种任务和信息。

AI Slide Editor by CubeOne

06/25 07:07

The editor PowerPoint should've shipped

中文介绍 「CubeOne」推出的AI幻灯片编辑器，旨在提供比传统PowerPoint更智能、更高效的编辑体验。它利用AI技术辅助用户创建和优化演示文稿。

SayCraft

06/24 15:58

Build a web app by talking through a meeting

中文介绍 「SayCraft」允许用户通过语音会议的方式构建网页应用。该工具旨在简化应用开发流程，让非技术用户也能通过自然语言描述需求，快速生成应用。

Polygraph

06/24 15:52

Let AI agents see cross repo and maintain session memory.

中文介绍 「Polygraph」允许AI智能体跨代码库进行查看，并保持会话记忆。这使得AI在处理复杂开发任务时能更好地理解上下文和关联信息，提升协作效率。

Poweful Agents inside Slack

06/25 06:39

中文介绍 视频展示了在 Slack 内部运行的强大 AI 代理。这些代理旨在自动化工作流程、提升团队协作效率，并可能执行复杂任务，从而简化日常办公操作。

AI Agents Just Changed Forever: GLM 5.2, Codex Skills, Claude & Cursor

06/21 21:30

中文介绍 视频探讨 AI 代理领域发生的重大变革。内容提及 GLM 5.2、Codex Skills、Claude 以及 Cursor 等关键技术与平台，预示着 AI 代理功能和应用模式的深刻演进。

9 AI Agent Skills To Get Ahead of 99% of People

06/18 10:27

How to build high quality internal tools with Retool

06/18 05:21

SpaceX Just Bought Cursor for $60B. It’s About to Take OVER.

06/17 05:31

Why the Government Just Killed Claude Fable 5

06/13 11:50

This is Why They Banned Claude Fable 5

06/13 02:28

Hermes Agent NEW Super-App and DeepSeek v4 Catches Up To Opus 4.8?

06/07 04:10

ChatGPT And Codex Are Merging (This Changes Everything)

06/04 06:17

Claude Opus 4.8: Everything You Need to Know

06/01 06:27

花 2 月写了一款软件，结果 claude code router 上线了一模一样的功能

06/26 16:48

4 回复 · 程序员 节点

apple 涨价，心态炸了，紧急补仓！

06/26 16:07

32 回复 · Apple 节点

iPhone16pro 有必要换 iPhone17pro 吗？

06/26 14:50

13 回复 · Apple 节点

MacBook 键盘泼上了咖啡，有没有抢救的方法

06/26 13:08

5 回复 · Apple 节点

使用 kkRepo 搭建 Python PyPI 私服

06/26 11:10

8 回复 · Python 节点

apple 全系涨价!

06/26 10:41

21 回复 · Apple 节点

Mac mini M4 也涨价了， 丐版的 16+256 的现在官网上 5999，我记得之前是 4499 吧，电商的 pdd 和狗东都搜不到这个产品了

06/26 10:05

28 回复 · Apple 节点

iOS 中的“团结”这个是啥？是中毒了么？

06/25 22:08

5 回复 · Apple 节点

 Apple 平板和电脑涨价

06/25 21:16

72 回复 · Apple 节点

请问下 neo 和 air m5 的购买建议

06/25 18:09

8 回复 · Apple 节点

Captcha proves you're human. HATCHA proves you're not

06/26 20:06

43 points · 46 comments

We All Depend on Open Source. We Will Defend It Together

06/26 13:40

269 points · 129 comments

Libre Barcode Project

06/26 11:12

196 points · 32 comments

What happened after 2k people tried to hack my AI assistant

06/26 10:29

216 points · 80 comments

Framework's 10G Ethernet module exposes USB-C's complexity

06/26 09:10

221 points · 113 comments

The Garbage Collection Handbook: The Art of Automatic Memory Management (2nd Ed) (2023)

06/26 07:10

171 points · 31 comments

The 'papers, please' era of the internet will decimate your privacy

06/26 05:44

817 points · 372 comments

Un-0: Generating Images with Coupled Oscillators

06/26 04:50

159 points · 39 comments

Om Malik has died

06/26 04:33

https://runtimewire.com/article/om-malik-taught-silicon-vall...

An oral history of Bank Python (2021)

06/26 04:14

137 points · 50 comments

Parallel Parentheses Matching

06/26 04:12

101 points · 13 comments

The Doorman's Fallacy in action

06/26 04:00

139 points · 196 comments

Show HN: OpenKnowledge – open source AI-first alternative to Obsidian/Notion

06/26 00:04

Hi HN, Nick here. We’re launching OpenKnowledge (https://openknowledge.ai/), a “what you see is what you get” markdown editor that has direct integrations with Claude, Codex, and other agents. Available as MacOS app or Web UI+CLI. Fully free/local and OSS.We built this because we

An entire Herculaneum scroll has been read for the first time

06/25 23:48

Preprint: https://scrollprize.org/pdf/main.pdfhttps://github.com/ScrollPrize/villa

IBM debuts sub-1 nanometer chip technology

06/25 23:33

339 points · 183 comments

OS9Map

06/25 23:01

238 points · 46 comments

Zig's new bitCast semantics and LLVM back end improvements

06/25 22:19

248 points · 125 comments

Apple raises prices of MacBooks, iPads

06/25 21:02

https://9to5mac.com/2026/06/25/apple-price-increases-mac-ipa...

Hey Nico, you didn't vibe code your data room but stole it from Papermark

06/25 20:23

https://xcancel.com/mfts0/status/2070080422482977095

You can't unit test for taste

06/24 16:54

288 points · 129 comments

Show HN: WebBase-III – dBASE III rebuilt in the browser with its own interpreter

06/24 16:46

7 points · 0 comments

Record type inference for dummies

06/23 21:04

52 points · 1 comments

A game where you're an OS and have to manage processes, memory and I/O events

06/23 17:35

251 points · 49 comments

Bipartite Matching Is in NC

06/23 06:43

23 points · 0 comments

Oxide computer 3D rack guided tour

06/22 23:21

398 points · 163 comments

The last Romans are still around

06/22 23:00

110 points · 149 comments

Doing a masters while working in Spain

06/22 15:44

62 points · 42 comments

Microbubbles in Medicine

06/22 09:59

15 points · 1 comments

Show HN: Chess-Inspired Roguelike

06/21 14:50

336 points · 111 comments

22-year-old Mozart's handwritten notebook unearthed in 'major discovery'

06/20 23:50

99 points · 21 comments