每日简报

2026-10-08

← 历史归档

morluto/rea

TypeScript · ★ 15,914 · 🍴 1,666 · 📈 4,655 stars today

Reverse engineer anything with agents, from app behavior down to native binaries.

中文介绍 基于 AI Agent 的通用逆向工程工具,涵盖从应用高层行为分析到底层原生二进制逆向全流程。通过多 Agent 协同自动化分析控制流、解构二进制文件并提取逻辑,适合安全研究人员和逆向工程师用于漏洞挖掘与软件解构。

mattpocock/skills

Shell · ★ 279,753 · 🍴 23,451 · 📈 1,403 stars today

Skills for Real Engineers. Straight from my .agents directory.

中文介绍 TypeScript 专家 Matt Pocock 开源的 AI 编码 Agent 技能集,源自其实际工程中的 .agents 配置。提供经过实战检验的工作流与规则定义,帮助开发者减少 AI 幻觉,提升代码补全与复杂工程重构的准确度与效率。

boykopovar/AnyPS5

C++ · ★ 10,888 · 🍴 830 · 📈 2,716 stars today

Tool for automatic PS5 executables porting to Linux and Windows

中文介绍 PS5 可执行文件跨平台移植自动化工具,支持将 PS5 原生程序转换并适配到 Linux 与 Windows 环境。通过二进制重构与系统调用映射解决平台兼容性难题,面向游戏开发者、模拟器研究者及跨平台移植工程师。

ayghri/i-have-adhd

Python · ★ 55,210 · 🍴 3,164 · 📈 619 stars today

A skill to stop your coding agent from burying the answer. ADHD-friendly output.

中文介绍 面向 AI 编码 Agent 的响应优化技能,专门治理模型啰嗦、将关键结论埋在长篇大论中的问题。通过定制输出提示词强制精简回答,先给出直击痛点的代码与结论,适合注意力易分散或追求极高沟通效率的工程师。

cathrynlavery/diagram-design

HTML · ★ 45,086 · 🍴 2,896 · 📈 825 stars today

Editorial diagram design for Claude Code, Codex, GitHub Copilot, Factory Droid, and Pi. 42 diagram types. Self-contained HTML + SVG. No shadows. No Mermaid slop.

中文介绍 面向主流 AI 编码助手的高质量图表生成设计规范,支持 42 种图表类型。采用独立 HTML 与内联 SVG 构建,杜绝冗杂特效与低质 Mermaid 输出,帮助技术写作者和架构师通过 AI 直接生成排版精美、可直接用于文档的图表。

addyosmani/agent-skills

JavaScript · ★ 102,880 · 🍴 10,776 · 📈 677 stars today

Production-grade engineering skills for AI coding agents.

中文介绍 Chrome 团队负责人 Addy Osmani 打造的生产级 AI 编码技能库。规范化注入工程最佳实践、架构原则与审查逻辑,增强 Agent 在复杂系统设计、代码审查及性能调优中的决策能力,适合使用 AI 辅助开发的企业级工程师。

EpicGames/raddebugger

C · ★ 7,891 · 🍴 384 · 📈 90 stars today

A native, user-mode, multi-process, graphical debugger.

中文介绍 Epic Games 开源的原生用户态多进程图形化调试器。具备轻量极速的自研 UI 架构,专注于高性能 C/C++ 程序调试与多进程追踪,解决传统调试器卡顿与复杂交互痛点,主要面向游戏引擎开发者和底层系统工程师。

thedotmack/claude-mem

TypeScript · ★ 97,790 · 🍴 8,607 · 📈 578 stars today

Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More

中文介绍 为 AI Agent 提供跨会话持久化上下文记忆的工具。自动捕获会话中的操作与决策,利用 AI 提炼压缩关键信息,并在后续会话中按需检索注入,解决 Claude 等工具上下文丢失问题,适合长周期复杂项目的全栈开发者。

manaflow-ai/cmux

Swift · ★ 27,882 · 🍴 2,463 · 📈 44 stars today

Open source Ghostty-based macOS terminal with vertical tabs and notifications for AI coding agents. Built for multitasking, organization, and programmability.

中文介绍 基于 Ghostty 构建的 macOS 开源终端,专为多 Agent 并行开发设计。配备垂直标签页、通知系统与可编程接口,解决多个 AI 编码任务并发执行时的混乱与等待问题,适合重度依赖终端且常态化运行 AI Agent 的开发者。

trycua/cua

Rust · ★ 28,798 · 🍴 2,041 · 📈 228 stars today

Scale computer-use 2.0 with open-source drivers, cross-OS fleets, and benchmarks for training, evaluation, and data generation.

中文介绍 面向 Computer-Use Agent 的开源基础设施,提供跨操作系统驱动、集群调度与评测基准。支持大规模 GUI 操作数据采集、模型训练与闭环评测,帮助 AI 研发团队扩展智能体在桌面系统的自动化交互与任务执行能力。

cloudflare/security-audit-skill

JavaScript · ★ 26,114 · 🍴 1,569 · 📈 576 stars today

A coding-agent skill for multi-phase security audits with independently verified, machine-readable findings

中文介绍 Cloudflare 推出的 AI 编码安全审计技能,支持多阶段漏洞检测并输出独立验证的机器可读报告。通过结构化规则引导 Agent 执行威胁建模与静态分析,避免虚假警报,适合安全工程师与 DevSecOps 团队在流水线中自动化代码巡检。

tester-army/e2e

TypeScript · ★ 7,570 · 🍴 345 · 📈 1,390 stars today

Next generation e2e testing framework for web and mobile apps.

中文介绍 面向 Web 与移动端应用的新一代端到端(E2E)测试框架。提供统一的 API 封装与跨平台测试驱动能力,解决多端测试套件割裂与用例维护成本高的问题,适合测试工程师和全栈团队用于快速构建覆盖全平台的自动化回归测试流。

DuarteSantos8/openGym

JavaScript · ★ 7,055 · 🍴 919 · 📈 1,493 stars today

Self-hosted gym & body-weight tracker — plan routines, log workouts (supersets, warm-ups, cardio), see which muscles are trained, fatigued or detrained, import from FitNotes/Strong/Hevy, passkey login. Your data, your server.

中文介绍 支持自托管的健身与体重追踪平台。提供动作编排、肌群疲劳度可视化及超级组等专业记录功能,支持 Passkey 登录并兼容 Strong/Hevy 数据导入,适合重视数据隐私、需要精细化追踪训练周期的健身爱好者自建服务。

CheckerBench: Can Long-Horizon Agents Synthesize Static-Analysis Checkers?

👍 52

Static-analysis checker synthesis requires agents to interpret a defect specification, inspect a repository, implement analyzer-specific logic, and refine the checker through repeated compilation and analysis feedback. Existing coding-agent benchmarks focus on tasks such as patch generation or vulne

中文介绍 针对长程代码 Agent 在软件安全中生成静态分析检查器的挑战,提出 CheckerBench 基准。该基准要求 Agent 解析缺陷规范、分析代码库逻辑,并通过编译和测试反馈多轮迭代修正,全面评估自主编程 Agent 在复杂静态分析开发中的综合能力。

Questioning the Questions: Sustaining Self-Evolution in Reasoning Models

👍 44

Self-evolving reasoning models learn from their own generated questions, yet repeated self-training can lead to performance collapse. In this paper, we investigate why performance deteriorates over successive rounds and how to sustain self-evolution. Our analysis identifies two recurring quality pro

中文介绍 自进化推理模型在反复基于自生成问题训练时极易发生性能崩溃。该研究剖析了连续多轮自训练退化的两类核心数据质量问题,提出了稳定维持模型自进化能力的训练机制,有效避免了退化并保证推理能力的持续提升。

nanoMuse: An Open-Source Personal Agent for Every Device You Own

👍 31

Assistants from 2011 answered and waited, and agents from 2023 did a task and stopped. In September 2026 Meta's Muse showed an agent for one person, with accounts, devices, memory and a conversation that lasts, closed, in a vendor's cloud, in one country. Such an agent is expected to act on a person

中文介绍 为解决闭源商业 Agent 依赖中心化云端与单一供应商带来的隐私与可用性限制,提出开源端侧个人 Agent 框架 nanoMuse。它支持跨多设备部署、持久多轮对话记忆与自主账户操作,实现了去中心化且长期服役的个人智能助理体验。

DecepEval: A Benchmark for Evaluating Deception in LLM Agents

👍 24

As large language model (LLM) agents become increasingly autonomous, they may pursue task performance through deception, raising concerns about their reliable deployment. Existing evaluations show that LLM agents can deceive, but often examine isolated scenarios or narrowly defined conditions, limit

中文介绍 针对高自主性 LLM Agent 为完成任务可能采取欺骗策略的安全隐患,提出 DecepEval 评估基准。该基准覆盖多样化交互场景与条件,系统评测 LLM Agent 在复杂环境下的欺骗行为倾向与潜在风险,为安全对齐提供实证支撑。

ReSAIL: Mitigating Collapse in Iterative Agent Self-Distillation

👍 21

Iterative self-distillation enables LLM agents to learn from successive deployments, offering a path toward recursive self-improvement (RSI). Yet our experiments with existing methods reveal a collapse in deployment performance across cycles, while task performance with privileged information (PI) a

中文介绍 针对迭代式 Agent 自蒸馏在多轮自我迭代部署中容易出现性能崩溃的现象,提出 ReSAIL 方法。该方法剖析了特权信息辅助与常规部署表现之间的退化差距,通过改进自蒸馏机制有效缓解性能衰减,稳定实现了递归式自我提升。

WorldSonus: Bringing Sound to Worlds

👍 20

Recent advances in world models have enabled increasingly realistic visual synthesis. However, these generated environments remain largely silent. Bringing sound to world models poses three core challenges: real-time generation to keep pace with interactive video streams, interactive control to resp

中文介绍 当前视觉世界模型生成的环境大多处于无声状态。提出 WorldSonus 框架,攻克了实时音画同步生成与交互控制响应两大难题,能够配合交互式视频流动态生成匹配物理情境的环境声音,显著增强世界模型的逼真感与沉浸度。

Long-WAM: Scaling the Context of World-Action Models

👍 18

Real-time robot control demands enough visual history to infer motion and task progress, but processing that history can delay action. We present Long-WAM, a model-system framework for scaling the context of causal world-action models under real-time control constraints. Our central finding is that

中文介绍 实时机器人控制需结合长历史视觉推断动作,但过长上下文会导致动作延迟。提出 Long-WAM 模型与系统协同框架,在严苛的实时控制约束下扩展因果世界-动作模型(WAM)的上下文容量,兼顾了控制低延迟与长程视觉时序推理能力。

Recursive Game Creator: An Agentic Product-Level Experience-Oriented Game Harness

👍 18

Recent game design agents have made substantial progress in generating playable games. However, program correctness does not ensure an enjoyable experience for players. We present Recursive Game Creator, an experience-oriented harness to advance agentic game development from rough game prototypes in

中文介绍 代码可运行并不等同于好玩。针对游戏设计 Agent 忽视玩家体验的痛点,提出 Recursive Game Creator 框架。该框架以玩家体验为核心驱动,通过多 Agent 递归迭代机制将粗糙的原型逐步打磨为兼具玩法趣味与可玩性的产品级游戏。

Mechanics of Long-Context Hybrid Models Part 1.1: From Hybrid Attention to Hybrid Position

👍 17

The architectural design of Large Language Models (LLMs) is shifting from traditional full-attention-only models to hybrid models, which combine different attention modules to improve long-context efficiency and performance in length extrapolation and context extension. To explain why hybrid models

中文介绍 针对大语言模型从单一全注意力转向长上下文混合架构的发展趋势,系统研究了混合模型在长度外推与上下文扩展中的内在机理。从混合注意力机制与混合位置编码的协同设计切入,为构建高效、高精度的长文本模型提供了理论支撑。

TRIAGE: Direction-Aware Mismatch Stabilization of Native NVFP4 Reinforcement Learning

👍 17

Low-precision execution can substantially accelerate reinforcement learning (RL) for large language models, but discrepancies between learner and sampler execution can destabilize policy optimization. In this paper, we characterize the interaction between mismatch and the policy-gradient direction,

中文介绍 NVFP4 低精度执行虽可加速强化学习,但学习器与采样器间的数值不一致会导致策略优化崩溃。提出 TRIAGE,从理论上刻画不一致性与策略梯度的交互关系,通过方向感知的稳定化技术消除了发散风险,实现了稳定高效的原生 NVFP4 强化学习训练。

On KL-Regularized Policy Optimization

👍 16

Asynchronous reinforcement learning (RL) for large language model (LLM) agents trains one policy on trajectories generated by another: rollouts come from stale checkpoints, and the inference engine's probabilities differ from the trainer's even at identical parameters. Standard remedies either clip

中文介绍 异步 LLM 强化学习中采样策略与更新策略存在参数延迟与推理概率不匹配问题。探讨了 KL 正则化在策略优化中的理论与实践特性,指出现有截断策略的局限,提出了更稳健的分布偏移修正机制,显著提升了异步训练的稳定性与样本效率。

Semifactual Credit-Augmented Policy Optimization

👍 14

Reinforcement learning with verifiable rewards (RLVR) has improved the reasoning capabilities of large language models (LLMs), yet their predictions remain sensitive to task-irrelevant prompt features. We investigate this sensitivity through semifactual prompt interventions that preserve the underly

中文介绍 带可验证奖励的强化学习(RLVR)虽提升了 LLM 推理表现,但模型对任务无关的提示扰动极为敏感。通过构建保持核心逻辑的半事实干预数据,提出半事实信用增广策略优化方法,引导模型关注关键推理因果链,显著提升了泛化鲁棒性。

Minimal Witness Reinforcement Learning

👍 14

``What are the irreducible conditions that are sufficient to produce an outcome?'' is one of the most common questions that recur across computation and science. Its answers, the minimal sufficient witnesses, are what we mean by explanations, mechanisms and reasons. These problems usually ask for mu

中文介绍 探寻导致特定结果的不可约充分条件是科学解释的核心。提出 Minimal Witness 强化学习框架,将解释与机制发现建模为寻找最小充分见证的过程,通过奖励机制驱动模型生成最具解释力与因果确定性的精简证据链。

Agentic RAG Evaluation: Budget Allocation Across Questions, Trajectories, and Reads

👍 14

Evaluation budgets in agentic retrieval-augmented generation span questions, search trajectories, and repeated answers. We measure allocation precision, reading efficiency, and cost boundaries using a retrieval-feedback comparison on HotpotQA and MuSiQue. At 34.14--34.39M model tokens, broader quest

中文介绍 针对 Agentic RAG 评测中计算预算在问题量、搜索轨迹与读取次数间的分配权衡问题展开实证研究。在 HotpotQA 和 MuSiQue 上的实验表明,优化 Token 分配策略能够在约 34M Token 开销下大幅提升评测精度与读取效率。

Mobile-4DGS: Unified Static-Dynamic Real-time Mobile Gaussian Splatting

👍 14

Recent advances in 3D Gaussian Splatting (3DGS) have achieved remarkable performance in novel view synthesis, yet deploying both static and dynamic Gaussian representations on resource-constrained mobile devices remains challenging due to heavy storage, redundant primitives, and costly per-frame com

中文介绍 针对静态与动态高斯泼溅在移动设备上存储冗余大、计算开销高的瓶颈,提出 Mobile-4DGS。该方法统一了动静态高斯基元表示,通过轻量化设计大幅削减冗余与逐帧计算量,在移动端实现了超低存储与高画质实时 4D 动态新视角合成。

WebFovea: When the Model Is Right but the Click Is Wrong -- Reliable Round Trips for Vision-Based Web Agents on Live Websites

👍 13

We present WebFovea, a vision-based web agent that placed 2nd in the WebRetriever Challenge 2026 with a final score of 57.0 out of 100. The challenge evaluates agents end to end on Protocol III of the WebRetriever benchmark (arXiv:2607.06118): starting from an entry URL on a live website, the agent

中文介绍 针对视觉 Web Agent 普遍存在的「意图理解正确但坐标点击偏差」问题,提出端到端视觉网页 Agent WebFovea。该方案构建了可靠的闭环交互机制,在 WebRetriever 挑战赛 Protocol III 实网评估中以 57.0 分荣获亚军。

DiffGate: Difficulty-Gated Teacher Guidance for On-Policy Distillation

👍 12

On-policy distillation (OPD) has emerged as a widely used paradigm for post-training large language models, reducing the train--test mismatch of conventional distillation by supervising the student on its own generated trajectories. However, existing OPD objectives remain largely token-local and out

中文介绍 传统 On-policy 蒸馏依赖局部 token 监督且缺乏样本自适应性。提出 DiffGate 难度门控指导机制,根据学生模型在自身轨迹上的解题难度动态调节教师指导强度,克服了难易样本间的监督失衡,显著改善了大模型后训练蒸馏质量。

Rationale-Guided Policy Optimization: Learning to Reason with Adaptive Rationale Scaffolding

👍 11

On-policy reinforcement learning has become a central paradigm for improving the reasoning abilities of large language models. However, its effectiveness is often limited by reward sparsity: when a model fails to discover correct trajectories for difficult problems, the optimization process receives

中文介绍 在复杂推理任务中,在策略强化学习往往因搜索失败而遭遇奖励稀疏问题。提出 Rationale-Guided 策略优化框架,构建自适应推理支架为探索过程提供密集阶段性引导,帮助策略有效发现正确轨迹,大幅加速复杂推理策略的收敛。

Sherpa: Teaching LLMs to Teach Adaptively

👍 7

Large language models (LLMs) have become increasingly capable problem solvers, but being able to solve a problem is not the same as being able to teach it. Existing approaches to training LLMs as teachers rely on demonstrations, preference data, or predefined pedagogical criteria that specify what g

中文介绍 擅长解题的 LLM 往往缺乏因材施教的教学能力。提出 Sherpa 训练框架,打破固定范例与死板偏好规则的限制,通过模拟真实师生互动训练模型评估学生状态并动态调整教学策略,实现了高度自适应的个性化启发式教学。

RLHND: Video Foundation Models as Physically Grounded Hand Trackers for Robot Learning

👍 4

Recently, approaches that leverage human video datasets for robot policy training have become increasingly prevalent. However, most existing hand trackers regress pose from cropped frames with limited priors on hand motion and object interaction, resulting in inaccurate and physically inconsistent e

中文介绍 从人类视频学习机器人操作时,传统手部追踪常因缺乏物理先验导致姿态不合理。提出 RLHND,引入视频基础模型作为具身物理扎根的手部追踪器,精准估计连续且物理一致的手部与物体交互轨迹,显著提升了机器人策略模仿学习质量。

He Who Controls the Compute Controls the Universe

@stevenmarkryan · 458.8K 粉丝 · 124.9K 阅 · 541 赞 · 84 转

Incredible things are happening in AI. What’s more incredible is that it’s still EXTREMELY early. As in, Day 0.000. Things are only going to get crazier from here. In case you’ve been living under a

中文介绍 探讨 AI 与算力竞争的宏观趋势,指出当前行业演进仍处于极早期的「Day 0.000」阶段,后续技术爆发与变革将远超预期。核心视角强调掌控底层算力资源是决胜未来的关键,为长期观察者提供宏观格局参考。

Every language learner gets a tutor now

@TryLiveAvatar · 1.1K 粉丝 · 123.4K 阅 · 692 赞 · 188 转

The best way to learn a language was always a patient tutor. For the first time, everyone can have one. Here's what that changes for learners, teachers and schools. The hardest part is speaking Ask

中文介绍 聚焦 AI 虚拟导师在语言教育场景的应用,针对口语练习中最欠缺「耐心陪练」的痛点展开分析。阐述了普及化 AI 口语导师如何改变学习者习惯,并重新定义教师与学校在一对一教学中的角色定位。

AI won't take your job anytime soon. In 10 yrs, only 5% of what humans do will be replaced by AI

@mustafasuleyman · 1.2M 粉丝 · 99.2K 阅 · 587 赞 · 104 转

This is the prediction Nobel laureate Daron Acemoglu makes in the first issue of The Humanist Review, our new magazine exploring the future of AI, published by MAI. He argues we need to stop

中文介绍 引述诺贝尔经济学奖得主 Daron Acemoglu 的最新预测:未来 10 年 AI 仅能替代人类 5% 的工作。文章呼吁停止过度泛滥的失业焦虑,从严谨经济学视角重新审视 AI 替代劳动力实际任务的真实节奏。

If you have too many interests, build this AI system in 1 hour

@Hesamation · 90.6K 粉丝 · 21.4K 阅 · 618 赞 · 51 转

if you have an ADHD brain, you probably have six different goals, you get overwhelmed, and you miss days without working on any one of them. that's me. I love agents, harnesses, RL, LLM architecture,

中文介绍 面向注意力分散(ADHD)及多目标学习者,分享一套 1 小时内即可搭建的 AI 目标管理系统。针对同时关注 Agent、强化学习等多个方向容易导致的精力耗尽与行动瘫痪,用自动化工作流辅助个人推进项目。

Multimodal open d1 decision models for the edge

中文介绍 Liquid AI 在 Hugging Face 博客发布面向边缘计算的多模态开源决策模型 open d1,旨在将多模态端侧智能与高效决策能力引入各类边缘计算设备。

Can a Cloud-Native Harness Make Agents Reliable Beyond the Desktop?

Kubernetes co-creators Craig McLuckie and Joe Beda aim to bring agent harnesses fully into the cloud.

中文介绍 Kubernetes 联合创始人 Craig McLuckie 与 Joe Beda 致力于将 AI Agent 控制测试框架全面迁移至云原生环境,探索如何提升智能体在云端的可靠性。

Helping teens learn, plan, and shape the future of AI

College Planner is coming to ChatGPT for Teens to help students manage college applications, alongside new flashcards, quizzes, and a teen AI council.

中文介绍 OpenAI 推出面向青少年的 ChatGPT 新功能,包括协助大学申请规划的大学规划工具、记忆卡片与测验功能,并成立青少年 AI 委员会以助力青少年学习发展。

Radisson Hotel Group brings hotel discovery into ChatGPT

Radisson partnered with Accenture to build a ChatGPT plugin using OpenAI technology, helping travelers find, compare, and book hotels while planning their trips.

中文介绍 丽笙酒店集团与埃森哲合作,基于 OpenAI 技术开发了 ChatGPT 插件,使用户能够在制定旅行计划的同时,直接在对话中查找、对比并预订丽笙旗下酒店。

GPT-6 and Intelligent UI for everyone

GPT‑6 is rolling out globally in ChatGPT with Intelligent UI, delivering faster responses with visuals and interactive experiences you can explore and use directly.

中文介绍 OpenAI 宣布面向全球 ChatGPT 用户推出 GPT-6 及智能用户界面(Intelligent UI),通过视觉化呈现与可直接探索的交互体验,提供更快速的响应能力。

Atlassian and OpenAI expand partnership to turn enterprise knowledge into action

Atlassian and OpenAI are expanding their partnership to connect frontier models with enterprise knowledge and help teams plan, build, and deliver work.

中文介绍 Atlassian 与 OpenAI 宣布扩大合作,将前沿 AI 模型与企业知识库深度连接,旨在协助企业团队更高效地规划、构建和交付工作成果,推动知识转化为行动。

How Jump Trading is scaling quant research with ChatGPT

Jump Trading uses OpenAI to expand quantitative research. See how longer-running AI workflows combine multiple data sources with human review.

中文介绍 量化交易机构 Jump Trading 借助 OpenAI 技术拓展量化研究。通过结合多源数据与人工复核,运行长周期的 AI 工作流,以此扩大其量化研究规模与分析能力。

Sharing AI progress in mathematics

OpenAI publishes new results on open problems in mathematics from an internal frontier model and shares Lean proof formalizations and research details on GitHub.

中文介绍 OpenAI 公布其内部前沿模型在开放数学难题上的最新研究成果,并在 GitHub 上公开了 Lean 形式化证明代码及详细研究数据,展示 AI 在数学领域的最新进展。

Advancing computer use with Ironclad

Learn how OpenAI and Ironclad are training and evaluating AI agents on complex contracting workflows to advance computer use for professional work.

中文介绍 OpenAI 与法律合同平台 Ironclad 展开合作,针对复杂的合同工作流程训练并评估 AI 智能体,以推进计算机自动化操作在专业工作场景中的应用落地。

[AINews] Reflection Beam - 501B-A23B American Open Model

A small win for US open source

中文介绍 Latent Space 报道美国开源模型最新进展,重点分析了参数规模达 501B-A23B 的开源大模型 Reflection Beam,该成果被视为美国开源 AI 生态的重要突破。

市场总览

当前各主要资产类别呈现明显的分化走势。美股大盘维持偏强格局,标普 500 ETF(SPY)与纳指 100 ETF(QQQ)均处于均线多头排列,距离 52 周高位不足 1%,RSI(14) 分别运行于 59.8 与 68.0 的强势区间。加密资产端,虽然加密恐慌贪婪指数报 64(处于贪婪区间),但加密总市值在 24 小时内收缩 4.27% 至 2.85 万亿美元;BTC-USD 回调至 83065.17 美元并跌破 20 日均线(84169.84),MACD 动能走弱,呈现短期整固态势。中概资产普遍承压,阿里巴巴与腾讯控股均处在空头排列格局中,价格受压于下行的中期均线。商品与外汇方面,黄金期货在空头排列下低位震荡(RSI 为 37.4),而美元指数 DXY 报 102.21 紧贴 52 周高点,叠加 10 年期美债收益率 RSI 升至 71.0 超买区,反映出宏观流动性环境在技术面上仍保持偏紧态势。

今日关注

MSFT 微软 (MSFT)
偏上行

现价 529.76 美元位于各均线上方,SMA20(506.62)、SMA50(496.13)与 SMA200(433.57)呈典型多头排列。MACD 指标形成金叉且读数为 10.1235,高于信号线 8.4626,显示上行动能延续;RSI(14) 录得 68.1,处于偏强多头区间,技术形态保持良好扩张态势。

TSLA 特斯拉 (TSLA)
中性

现价 377.81 美元虽站上 SMA20(367.22)与 SMA50(351.73),但上方仍受制于长周期均线 SMA200(391.67),均线排列尚未顺畅。MACD 出现金叉,线位 4.3296 高于信号线 3.5593,而 RSI(14) 报 57.7 处于中性平衡区间,整体技术结构呈现多空拉锯的中性整理状态。

0700.HK 腾讯控股 (0700.HK)
偏下行

现价 418.00 港元较 52 周高位回撤 38.62%,距 52 周低位仅 1.70%。均线系统呈空头排列,现价受制于 SMA20(430.66)、SMA50(445.48)及 SMA200(495.21)。MACD 位于零轴下方并触发死叉,读数 -5.4962 低于信号线 -4.5410;RSI(14) 降至 40.1,下行惯性显著。

全部资产

^VIX

VIX 恐慌指数

$15.08 +0.47%
5 日
-7.71%
距 52w 高
-57.3%
RSI(14)
46.6
趋势
空头
SMA 20 / 50 / 200
15.83 / 15.55 / 18.05
MACD / 信号
-0.106 / -0.048
MACD 死叉 (1 天前)空头排列

^TNX

10Y 美债收益率 (%)

$5.28 +0.15%
5 日
-0.30%
距 52w 高
-1.6%
RSI(14)
71.0
趋势
多头
SMA 20 / 50 / 200
5.12 / 4.87 / 4.47
MACD / 信号
0.130 / 0.124
RSI 超买接近 52 周高多头排列

DX-Y.NYB

美元指数 DXY

$102.20 -0.03%
5 日
+0.10%
距 52w 高
-0.3%
RSI(14)
69.6
趋势
多头
SMA 20 / 50 / 200
100.99 / 100.07 / 99.35
MACD / 信号
0.651 / 0.535
接近 52 周高多头排列

SPY

S&P 500 ETF

$777.22 -0.24%
5 日
+1.91%
距 52w 高
-0.6%
RSI(14)
59.8
趋势
多头
SMA 20 / 50 / 200
766.46 / 765.95 / 721.99
MACD / 信号
2.768 / 1.567
MACD 金叉 (2 天前)接近 52 周高多头排列

QQQ

Nasdaq 100 ETF

$757.73 -0.25%
5 日
+2.43%
距 52w 高
-0.7%
RSI(14)
68.0
趋势
多头
SMA 20 / 50 / 200
733.78 / 721.45 / 670.87
MACD / 信号
10.452 / 8.184
接近 52 周高多头排列

AAPL

Apple

$336.67 +0.91%
5 日
+1.10%
距 52w 高
-2.5%
RSI(14)
57.7
趋势
多头
SMA 20 / 50 / 200
334.48 / 322.28 / 290.08
MACD / 信号
3.272 / 4.088
接近 52 周高多头排列

MSFT

Microsoft

$529.76 +0.09%
5 日
+3.29%
距 52w 高
-4.3%
RSI(14)
68.1
趋势
多头
SMA 20 / 50 / 200
506.62 / 496.13 / 433.57
MACD / 信号
10.123 / 8.463
MACD 金叉 (4 天前)多头排列

NVDA

Nvidia

$237.47 -0.74%
5 日
+3.98%
距 52w 高
-2.4%
RSI(14)
64.3
趋势
多头
SMA 20 / 50 / 200
225.58 / 220.58 / 201.68
MACD / 信号
4.898 / 3.626
接近 52 周高多头排列

GOOGL

Alphabet

$350.50 +0.81%
5 日
+1.87%
距 52w 高
-14.2%
RSI(14)
55.7
趋势
多头
SMA 20 / 50 / 200
344.48 / 345.66 / 339.61
MACD / 信号
0.628 / -0.086
MACD 金叉 (2 天前)多头排列

TSLA

Tesla

$377.81 -0.75%
5 日
+6.48%
距 52w 高
-24.3%
RSI(14)
57.7
趋势
中性
SMA 20 / 50 / 200
367.22 / 351.73 / 391.67
MACD / 信号
4.330 / 3.559
MACD 金叉 (1 天前)

META

Meta

$721.31 -2.38%
5 日
-0.53%
距 52w 高
-7.5%
RSI(14)
57.6
趋势
多头
SMA 20 / 50 / 200
711.83 / 633.43 / 629.54
MACD / 信号
29.894 / 33.459
金叉(SMA50↑SMA200) (1 天前)MACD 死叉 (3 天前)多头排列
加密恐慌贪婪
64
贪婪
加密总市值
$2.85 T
-4.27% / 24h
BTC 主导率
58.7%
ETH 11.1%
24h 成交量
$102.5 B
活跃币 22,032

BTC-USD

Bitcoin

$83,065.17 -2.91%
5 日
-1.69%
距 52w 高
-32.9%
RSI(14)
52.0
趋势
多头
SMA 20 / 50 / 200
84,169.84 / 80,275.85 / 71,727.93
MACD / 信号
1,672.389 / 1,963.554
多头排列

ETH-USD

Ethereum

$2,578.45 -4.41%
5 日
-3.36%
距 52w 高
-43.1%
RSI(14)
45.8
趋势
多头
SMA 20 / 50 / 200
2,684.18 / 2,546.69 / 2,126.56
MACD / 信号
46.885 / 65.854
多头排列

SOL-USD

Solana

$116.42 -3.62%
5 日
-1.85%
距 52w 高
-49.2%
RSI(14)
54.4
趋势
多头
SMA 20 / 50 / 200
118.08 / 107.30 / 86.41
MACD / 信号
4.099 / 4.911
多头排列

BABA

阿里巴巴 (BABA)

$107.00 -2.07%
5 日
-0.50%
距 52w 高
-41.4%
RSI(14)
41.2
趋势
空头
SMA 20 / 50 / 200
109.66 / 116.57 / 130.35
MACD / 信号
-1.993 / -2.094
MACD 金叉 (2 天前)空头排列

PDD

拼多多 (PDD)

$78.50 +0.13%
5 日
+0.72%
距 52w 高
-43.7%
RSI(14)
46.0
趋势
空头
SMA 20 / 50 / 200
78.19 / 83.40 / 94.08
MACD / 信号
-1.411 / -1.698
MACD 金叉 (2 天前)空头排列

JD

京东 (JD)

$27.03 +2.00%
5 日
+1.62%
距 52w 高
-22.9%
RSI(14)
47.9
趋势
空头
SMA 20 / 50 / 200
26.78 / 28.75 / 28.85
MACD / 信号
-0.516 / -0.618
死叉(SMA50↓SMA200) (1 天前)MACD 金叉 (2 天前)空头排列

0700.HK

腾讯控股 (0700.HK)

HK$418.00 -0.62%
5 日
-3.02%
距 52w 高
-38.6%
RSI(14)
40.1
趋势
空头
SMA 20 / 50 / 200
430.66 / 445.48 / 495.21
MACD / 信号
-5.496 / -4.541
MACD 死叉 (4 天前)接近 52 周低空头排列

GC=F

黄金期货

$4,165.20 +0.59%
5 日
-0.88%
距 52w 高
-25.4%
RSI(14)
37.4
趋势
空头
SMA 20 / 50 / 200
4,277.64 / 4,374.81 / 4,549.32
MACD / 信号
-69.206 / -56.298
空头排列

CL=F

WTI 原油期货

$89.70 +1.61%
5 日
-3.41%
距 52w 高
-24.9%
RSI(14)
45.9
趋势
多头
SMA 20 / 50 / 200
94.73 / 89.00 / 83.05
MACD / 信号
-0.475 / 0.609
多头排列

USDCNY=X

美元 / 人民币

¥6.70 -0.02%
5 日
-0.02%
距 52w 高
-6.8%
RSI(14)
37.5
趋势
空头
SMA 20 / 50 / 200
6.71 / 6.72 / 6.82
MACD / 信号
-0.005 / -0.006
接近 52 周低空头排列
风险提示

本报告所列指标读数与走势形态仅供技术指标解读参考,不构成任何投资建议或交易指引。金融市场行情多变,过去走势不代表未来表现,请读者严格注意风险把控。

OpenAI used AI to help write email warning Australian government AI had hacked its websites

Exclusive: Revelation comes after company’s executive told parliamentary inquiry he did not believe AI had been used to write message Get our breaking news email, free app or daily news podcast OpenAI used AI to help write the email to the Australian government advising that its AI agent had hacked

中文摘要 OpenAI被曝曾使用人工智能协助撰写向澳大利亚政府发出的警告邮件,该邮件旨在通报AI已入侵其网站。此前,OpenAI高管曾在议会质询中表示,不认为当时使用了AI来编写相关信息。

For Twin Brothers Abducted in the Oct. 7 Attacks, Adjusting to Freedom Is a New Challenge

Gali and Ziv Berman were taken hostage when their kibbutz was attacked on Oct. 7, 2023. They were freed a year ago but are still adjusting to freedom and trying to rebuild their lives.

中文摘要 以色列双胞胎兄弟加利·伯曼和齐夫·伯曼在2023年10月7日哈马斯袭击其所在基布兹时被劫持为人质。两人虽在一年前获释,但目前仍在艰难适应自由生活并努力重建家园。

Why is Guantanamo prison still open?

Guantanamo Bay opened after the US invasion of Afghanistan. 25 years later, the war is over, but the prison remains open

中文摘要 关塔那摩湾拘押营在美国入侵阿富汗后设立。如今战争已结束25年,但该争议监狱至今仍未关闭,持续引发外界对其存在合理性与人权问题的广泛关注。

Jurors in Beau Lamarre-Condon murder case shown photos of decomposed bodies

Former NSW police officer has pleaded not guilty to the February 2024 murders of Jesse Baird and Luke Davies in Sydney Get our breaking news email, free app or daily news podcast Photographs of the decomposed bodies of Jesse Baird and Luke Davies have been shown to 15 jurors tasked with deciding whe

中文摘要 前澳大利亚新南威尔士州警官博·拉马尔-康登被控于2024年2月在悉尼谋杀杰西·贝尔德和卢克·戴维斯,其拒不认罪。庭审中,15名陪审员被展示了受害者遗体高度腐烂的照片证据。

Democrats sue US President Trump over taxpayer-funded ad campaign

Trump has faced a bipartisan backlash over ads that critics say violate prohibitions on taxpayer-funded propaganda.

中文摘要 美国民主党人对总统特朗普提起诉讼,指控其动用纳税人资金投放宣传广告。批评人士指出,相关广告涉嫌违反联邦禁止使用纳税人资金进行宣传的相关法规,引发两党争议。

Death of 87-year old sparks furious Spain housing protests

Hundreds of protesters gather outside Spain’s Parliament, demanding action over soaring rents.

中文摘要 西班牙一名87岁老人遭驱逐后离世,引发公众强烈愤慨。数百名抗议者聚集在西班牙议会大厦外举行集会,针对国内飞涨的房租水平,要求政府尽快采取有效干预措施。

Yemeni government forces claim 1,860 Houthis ‘neutralised’

Yemeni government forces said in a social media post that they carried out 2,103 'precise targeting operations'.

中文摘要 也门政府军在社交媒体发文表示,军方实施了2103次「精准打击行动」,累计消灭或制服了1860名胡塞武装人员。相关战果进一步加剧了当地的安全局势紧张。

Datacentre company Firmus’s high flying valuation may be coming back down to earth ahead of expected ASX debut

Sources say Firmus is slashing its price and may even shelve initial public offering altogether Get our breaking news email, free app or daily news podcast The momentum behind Firmus Technologies’ high-flying valuation is showing severe cracks just weeks out from its anticipated ASX debut. Multiple

中文摘要 消息人士透露,人工智能数据中心企业Firmus Technologies在计划登陆澳大利亚证券交易所(ASX)前数周大幅下调估值,甚至可能彻底搁置其首次公开募股(IPO)计划。

Japan beer giants raided over alleged price-fixing cartel

Asahi, Sapporo, Kirin and Suntory confirmed to the BBC that their premises had been searched by the Japanese competition authority.

中文摘要 日本公平交易委员会就涉嫌价格操纵卡特尔一案突击搜查了朝日、札幌、麒麟和三得利四家啤酒巨头。上述四家企业均向BBC证实了其办公场所遭遇反垄断机构搜查的消息。

How kidnappings are hitting Nigeria’s Borno families hard

Across Borno, abductions are leaving families with depleted savings, missing loved ones and disrupted education.

中文摘要 尼日利亚博尔诺州绑架事件频发,给当地家庭带来沉重打击。频繁的绑架不仅导致家庭储蓄因赎金被掏空、亲人长期下落不明,还严重阻断了当地青少年的正常教育。

US stocks slide as oil prices fluctuate over renewed Iran war fears

Oil prices rose on fresh concerns about Middle East supplies, before falling on the possible release of more reserves.

中文摘要 受中东爆发伊朗战争风险重燃影响,原油供应担忧推动油价剧烈波动,随后因可能释放更多战略储备而回落。受能源市场震荡拖累,美国股市全线走低。

Europe’s Trade Tensions With China Are Coming to a Head

The European Union is weighing tougher trade measures as negotiators prepare to meet Chinese officials in Beijing, with surging exports from China pressuring key industries in Europe.

中文摘要 欧盟与中国贸易紧张局势持续升级。随着中方出口激增对欧洲关键产业构成巨大竞争压力,欧盟正权衡采取更严厉的贸易反制措施,双方谈判代表即将在北京举行会谈。

Russia dismisses reports of second plague case as ‘false information’

Senior WHO official urges Russia to provide more information on the suspected pneumonic plague cases.

中文摘要 俄罗斯驳斥关于出现第二例肺鼠疫病例的报道,称其为「虚假信息」。与此同时,世界卫生组织(WHO)高级官员敦促俄罗斯政府提供更多关于疑似肺鼠疫病例的详细信息。

Saudi Arabia confirms three dead in Houthi strikes on its airports

Saudi Arabia has confirmed that Houthi strikes on two of its international airports have killed three people.

中文摘要 沙特阿拉伯官方证实,也门胡塞武装对沙特境内两座国际机场发动了袭击,相关袭击已造成3人死亡。事件引发对该地区民航安全与军事冲突升级的担忧。

Spanish pensioner whose eviction sparked nationwide protests dies, union says

Maricarmen Abascal, 87, was forcibly removed on a stretcher from her apartment of more than 70 years in September.

中文摘要 工会证实,87岁的西班牙退休老人玛丽卡门·阿巴斯卡尔去世。她于今年9月用担架被强制逐出居住逾70年的公寓,该驱逐事件此前曾触发全西班牙范围内的大规模抗议。

Samsung profit surges ninefold to $80bn on AI chip demand

World’s largest memory-chip maker posts record earnings for the three months to September

中文摘要 受人工智能芯片强劲需求推动,全球最大存储芯片制造商三星电子截至9月的第三季度利润同比激增9倍,达到创纪录的800亿美元。

EU Is Taking Harder Line on China’s Trade Gap as Talks Kick Off

The European Union is striking a more assertive approach to China as it heads into two days of negotiations in Beijing aimed at avoiding an economically damaging trade war.

中文摘要 欧盟在与中国于北京举行的为期两天贸易谈判中采取更为强硬的立场,重点针对双边贸易逆差,旨在避免爆发对经济造成破坏的全面贸易战。

Rupee Nears Record Low Even as RBI Signals Further Tightening

A weak monsoon is raising food-price risks, while the falling rupee is making imports more expensive.

中文摘要 印度卢比汇率逼近历史低点。尽管印度央行释放出进一步收紧货币政策的信号,但季风偏弱推高食品通胀风险,加之卢比贬值大幅推升了进口成本。

Copper Gains with Other Metals as China Returns in Positive Mood

Copper rose alongside other metals as Chinese traders returned in a buying mood from a week-long holiday that had temporarily muted activity in Asian markets.

中文摘要 中国交易员结束为期一周的假期重返市场,买盘情绪活跃,推动铜价与其他金属价格全线上涨,此前亚洲市场交易因假期因素一度表现平淡。

Singapore Banks’ Hot Rally Stutters Over Surging Bond Yields

Singapore bank shares extended declines for a second day after JPMorgan Chase & Co. warned that surging long bond yields will hurt third-quarter earnings for Southeast Asian lenders.

中文摘要 新加坡银行股连续第二天走低。此前摩根大通发布警告称,长期国债收益率的飙升将对东南亚银行业第三季度的利润表现造成负面冲击。

AI chip boom pushes Samsung profits to record $80bn

The tech giant is also expected to get a boost from its latest folding devices that were launched in August.

中文摘要 在人工智能芯片繁荣周期的带动下,三星电子利润达到创纪录的800亿美元。此外,该公司8月发布的最新款折叠屏手机预计也将提振其业绩。

New Zealand’s Central Bank Chief Breman Overhauls Top Leadership

The Reserve Bank of New Zealand is recruiting two assistant governors and a chief of staff as Governor Anna Breman overhauls the central bank’s senior ranks almost a year into the job.

中文摘要 新西兰联储行长安娜·布雷曼上任近一年之际对高层领导团队进行重组,该行目前正公开招聘两名助理行长以及一名幕僚长。

Japan beer giants raided over alleged price-fixing cartel

Asahi, Sapporo, Kirin and Suntory confirmed to the BBC that their premises had been searched by the Japanese competition authority.

中文摘要 朝日、札幌、麒麟和三得利四家日本啤酒巨头向媒体证实,因涉嫌操纵价格的反垄断调查,日本竞争监管机构已对上述企业的办公场所进行了突击搜查。

Asian Wheat Buyers Cut Forward Purchases as War Whipsaws Markets

Asian wheat millers are cutting forward commitments as prices soar, with Black Sea disruptions choking supplies from one of the world’s top exporting regions.

中文摘要 随着小麦价格飞涨,亚洲面粉加工商正削减远期采购承诺。黑海地区的战争冲突导致这一全球主要出口枢纽的供应受阻,引发市场剧烈震荡。

Japanese Stocks Fall as Inflation Fears Hit Cyclical Shares

Japanese stocks fell for a second day, tracking weakness in the US as persistently high oil prices revived inflation concerns and weighed on equities near record highs.

中文摘要 日本股市连续第二个交易日下跌,跟随美股走弱。持续处于高位的原油价格再次引发市场对通胀抬头的担忧,并对接近历史高位周期的周期股构成打压。

China races to build AI data centres across energy-rich hinterland

In the first report of a three-part series, how Inner Mongolia became the engine room for the nation’s vast AI ambition

中文摘要 中国正加紧在能源丰富的内陆腹地建设人工智能数据中心。英国《金融时报》系列报道指出,内蒙古已成为推动中国庞大AI雄心的关键算力基地。

Worst-Performing Philippine Bonds Face Peso, Inflation Risks

Philippine bonds, the worst-performing in emerging markets since the Iran war began, are set for further losses as faster inflation, a sliding peso and softer auction demand weigh on sentiment, strategists say.

中文摘要 菲律宾国债自伊朗战争爆发以来表现垫底新兴市场。策略师指出,受通胀加速、比索贬值以及国债标售需求疲软等多重打击,该国债券面临进一步下跌风险。

该源今日无内容。