Cheng1937(∞) (@Cheng19371) profile picture
Taipei, Taiwan
Joined March 2021
Photo by Cheng1937(∞) (@Cheng19371)Photo by Cheng1937(∞) (@Cheng19371)Photo by Cheng1937(∞) (@Cheng19371)Photo by Cheng1937(∞) (@Cheng19371)Photo by Cheng1937(∞) (@Cheng19371)Photo by Cheng1937(∞) (@Cheng19371)
Cheng1937(∞) retweeted
Introducing GitHub sync on v0: • Push generated code to GitHub directly from v0 • v0 will automatically pull changes from GitHub into your chat • Switch branches and open PRs to collaborate with your team
66
145
71
1,627
270,514
Cheng1937(∞) retweeted
一个专为中文用户打造的 MCP(模型上下文协议)资源合集 Awesome-MCP-ZH github.com/yzfly/Awesome-MCP…
1
63
289
23,093
Cheng1937(∞) retweeted
使用 Cursor 不知道用哪些 MCP?文章推荐了 Cursor 中 4 个实用的 MCP 服务,我都安装上了,简单总结如下: 1. Sequential Thinking 核心能力:能把复杂问题分解为简单的步骤,便于 AI 分步处理,同时展示 AI 的思考过程。 使用场景:适合处理编程任务等复杂问题,逐步分析并优化解决方案。 2. Server Memory 核心能力:帮助 AI 记住此前交互内容,让对话更连贯。 使用场景:适合需求持续变化、多轮交互的编程和需求分析等任务,提升信息连续性和准确性。 3. Fetch 核心能力:从外部数据源获取信息,以 Markdown 格式返回,增强 AI 的信息准确性。 使用场景:适合需要外部数据支撑的任务,如专业知识查询和信息获取。 4. Playwright 和 Puppeteer 核心能力:实现浏览器自动化交互,包括网页操作、截图和执行 JavaScript 脚本。 使用场景:适合网页自动化测试和截图分析,节省大量人工操作时间。 这些 MCP 工具都通过 Cursor 简单配置和调用,能够显著提高日常开发和信息处理效率,推荐试试。
10
105
2
439
43,368
Cheng1937(∞) retweeted
Web search is now available on our API. Developers can augment Claude's comprehensive knowledge with up-to-date data.
69
208
76
1,677
214,210
Cheng1937(∞) retweeted
Google Agent Companion 白皮书 重读 Google 智能体指南白皮书,强调了 AgentOps、指标驱动、自动化评估结合人工反馈、多智能体架构、Agentic RAG、底层搜索优化、安全性的重要性 核心概念: Agent 能感知环境、推理、使用工具并自主行动以达成目标。核心组件包括模型 (Model)、工具 (Tools) 和编排层 (Orchestration Layer) AgentOps: 指 Agent 的运维实践,它继承了 DevOps/MLOps 的原则,并增加了对工具管理、编排、记忆和任务分解的关注。指标驱动的改进是关键 成功指标: 需要跟踪业务指标、目标完成率、关键任务成功率、延迟、错误,并结合人类反馈。详细追踪 (Trace) 用于调试 智能体评估: 需评估其核心能力、行动轨迹 (Trajectory) 和最终响应。方法包括基准测试、自动化轨迹分析、自动评估器 (Autorater) 以及不可或缺的人工评估 (Human-in-the-Loop) 多智能体系统: 由多个专业化智能体协作,可提高准确性、效率、可扩展性和容错性。挑战在于协调、沟通和任务分配。评估需关注协作效果 Agentic RAG: 一种更高级的 RAG,利用智能体 主动优化检索过程,通过查询扩展、多步推理、源选择和验证来提升效果。优化基础搜索是前提 企业级智能体: “助手型”(交互)和“自动化型”(后台)是主要形式。人类将成为“智能体管理者”。Google Agentspace 等平台支持这一趋势 智能体即“承包商” (Contractor): 提出通过标准化的“合同”来定义智能体 任务,明确交付物、范围、成本、验证标准和协商机制,以提高复杂任务的可靠性。支持通过子合同分解任务 汽车 AI 案例: 展示了多智能体 系统如何通过专业分工(导航、媒体等)和协作模式(层级、协作等)提升车载体验 核心启示: 强调 AgentOps、指标、评估、多智能体优势、Agentic RAG、安全性及“承包商”概念的重要性。鼓励利用现有工具进行开发
Google literally dropped a practical guide to building, deploy and evaluate multi-agent apps. It talks about AgentOps, Agentic RAG architecture and how you can build Google quality search using Vertex AI. As better search leads to better RAG.
9
72
2
245
45,691
Cheng1937(∞) retweeted
在尝试使用 MCP 服务器时,很多人都遇到过安装配置繁琐、环境依赖复杂的问题,特别需要同时管理多个 MCP 服务器,更加头疼。 开源项目 MCP Containers 试图改变这一现状,它将数百个常用 MCP 服务器打包成即用即取的 Docker 镜像,直接一键部署各种 AI 能力扩展的 MCP 工具。 GitHub:github.com/metorial/mcp-cont… 主要功能: - 一键部署体验,通过简单的 Docker 命令即可使用任何 MCP 服务器 - 自动更新机制,服务器镜像每日更新,始终与源仓库同步 - 安全隔离环境,各服务器在独立容器中运行互不干扰 - 支持数百种 MCP 服务器,从网络搜索、浏览器控制到各类 API 集成 - 丰富的文档和使用指南,帮助快速上手 只需拉取需要的 Docker 镜像即可,同时每一个 MCP 服务器都有详细的使用教程。
5
51
206
14,166
Cheng1937(∞) retweeted
假如你对 Web 实时同学感兴趣,可以看看这个「给好奇者的 WebRTC」,挺适合入门者学习。 webrtcforthecurious.com/zh/
14
42
256
24,135
OpenAI has published a new prompting guide for GPT-4.1 Agentic prompt that OpenAI used to achieve its highest score on SWE-bench Verified is added here and blog link in 2nd post. --------------------------------------------------------- You will be tasked to fix an issue from an open-source repository. Your thinking should be thorough and so it's fine if it's very long. You can think step by step before and after each action you decide to take. You MUST iterate and keep going until the problem is solved. You already have everything you need to solve this problem in the /testbed folder, even without internet connection. I want you to fully solve this autonomously before coming back to me. Only terminate your turn when you are sure that the problem is solved. Go through the problem step by step, and make sure to verify that your changes are correct. NEVER end your turn without having solved the problem, and when you say you are going to make a tool call, make sure you ACTUALLY make the tool call, instead of ending your turn. THE PROBLEM CAN DEFINITELY BE SOLVED WITHOUT THE INTERNET. Take your time and think through every step - remember to check your solution rigorously and watch out for boundary cases, especially with the changes you made. Your solution must be perfect. If not, continue working on it. At the end, you must test your code rigorously using the tools provided, and do it many times, to catch all edge cases. If it is not robust, iterate more and make it perfect. Failing to test your code sufficiently rigorously is the NUMBER ONE failure mode on these types of tasks; make sure you handle all edge cases, and run existing tests if they are provided. You MUST plan extensively before each function call, and reflect extensively on the outcomes of the previous function calls. DO NOT do this entire process by making function calls only, as this can impair your ability to solve the problem and think insightfully. # Workflow ## High-Level Problem Solving Strategy 1. Understand the problem deeply. Carefully read the issue and think critically about what is required. 2. Investigate the codebase. Explore relevant files, search for key functions, and gather context. 3. Develop a clear, step-by-step plan. Break down the fix into manageable, incremental steps. 4. Implement the fix incrementally. Make small, testable code changes. 5. Debug as needed. Use debugging techniques to isolate and resolve issues. 6. Test frequently. Run tests after each change to verify correctness. 7. Iterate until the root cause is fixed and all tests pass. 8. Reflect and validate comprehensively. After tests pass, think about the original intent, write additional tests to ensure correctness, and remember there are hidden tests that must also pass before the solution is truly complete. Refer to the detailed sections below for more information on each step. ## 1. Deeply Understand the Problem Carefully read the issue and think hard about a plan to solve it before coding. ## 2. Codebase Investigation - Explore relevant files and directories. - Search for key functions, classes, or variables related to the issue. - Read and understand relevant code snippets. - Identify the root cause of the problem. - Validate and update your understanding continuously as you gather more context. ## 3. Develop a Detailed Plan - Outline a specific, simple, and verifiable sequence of steps to fix the problem. - Break down the fix into small, incremental changes. ## 4. Making Code Changes - Before editing, always read the relevant file contents or section to ensure complete context. - If a patch is not applied correctly, attempt to reapply it. - Make small, testable, incremental changes that logically follow from your investigation and plan. ## 5. Debugging - Make code changes only if you have high confidence they can solve the problem - When debugging, try to determine the root cause rather than addressing symptoms - Debug for as long as needed to identify the root cause and identify a fix - Use print statements, logs, or temporary code to inspect program state, including descriptive statements or error messages to understand what's happening - To test hypotheses, you can also add test statements or functions - Revisit your assumptions if unexpected behavior occurs. ## 6. Testing - Run tests frequently using `!python3 run_tests.py` (or equivalent). - After each change, verify correctness by running relevant tests. - If tests fail, analyze failures and revise your patch. - Write additional tests if needed to capture important behaviors or edge cases. - Ensure all tests pass before finalizing. ## 7. Final Verification - Confirm the root cause is fixed. - Review your solution for logic correctness and robustness. - Iterate until you are extremely confident the fix is complete and all tests pass. ## 8. Final Reflection and Additional Testing - Reflect carefully on the original intent of the user and the problem statement. - Think about potential edge cases or scenarios that may not be covered by existing tests. - Write additional tests that would need to pass to fully validate the correctness of your solution. - Run these new tests and ensure they all pass. - Be aware that there are additional hidden tests that must also pass for the solution to be successful. - Do not assume the task is complete just because the visible tests pass; continue refining until you are confident the fix is robust and comprehensive.
43
227
20
1,655
229,741
Cheng1937(∞) retweeted
又看了一些推友的推荐: 还有这个 ArkType 验证库可以尝试: arktype.io/ 和 Zod 非常类似,但是速度非常快,语法简约更接近 Typesript 的原生写法,错误信息设置的更加清晰。 看了几个例子觉得非常不错。如果不是我的代码都和 zod 深度绑定,我还真想换换看。新项目可以尝试一下。 一个小结论,如果要做验证,那么目前就是这三个方案最棒: Zod v4.zod.dev/:最成熟/最流行 ArkType:更贴近原生语法,更快 Valibot valibot.dev/:轻量级,模块化,体积小。
前端大新闻,Zod 推出了新版本 Zod4, v4.zod.dev/v4 不仅仅是速度的提升,还添加了好多新特性,我比较感兴趣的几个: JSON Schema:主要是为了适配现在大模型 structured outputs 功能,可以完成一键转换,非常方便。 zod/mini:支持 tree shaking API,可以大幅减小打包以后的体积,简直是前端的福音了。 总体了说,对于我这个 zod 重度使用者来说,是非常值得升级的,如果还有朋友没有使用 zod 的话,但是有验证的需求,那么必需尝试一下。
10
16
75
23,215
We’ve open-sourced our MCP (Model Context Protocol) — now available on GitHub. Plug it into your favorite client and start building richer AI integrations with Notion in minutes ✨ github.com/makenotion/notion…
31
350
58
2,756
471,221
Cheng1937(∞) retweeted
一份精心整理的检索增强生成(RAG)技术资源集合:Awesome-RAG。 这份列表系统整理了 RAG 相关的理论知识、实现方法、开发框架、关键技术以及评估指标等内容。 GitHub:github.com/Danielskry/Awesom… 从基本原理到高级技术详解,再到实现步骤,有助于我们更好学习与使用 RAG。
3
53
1
186
12,290
Cheng1937(∞) retweeted
Meet Firebase Studio: A cloud-based, agentic dev environment powered by Gemini ✨💻✨ Find everything you need to prototype, build, and run production-quality full-stack AI apps quickly and safely. Learn more about building AI apps with Firebase → goo.gle/4j3MS9v #GoogleCloudNext
153
757
313
4,874
803,569
Cheng1937(∞) retweeted
牛P了 兄弟们 Google 推出类似 MCP 的 Agent2Agent 协议 (A2A) 可让 多个 Agent 跨平台互通并协作完成任务 组成一个智能协同体 也就是可以“让不同公司的 AI Agent 能像人类一样互相合作、对话、分工完成任务。” 多个 Agent 可以像流水线一样协作完成一个复杂任务。 你可以把它想象成是: 🧠 AI 代理之间的“通用语言” 🌐 类似 HTTP 是网页通信协议,A2A 是 Agent 之间通信的协议 👷‍♂️ 专门为多个 Agent 协同工作而生 A2A 与 MCP 有啥区别? 👉 A2A 是 Agent 之间对话、沟通、协作协议 👉 MCP 是 Agent 调用各种“工具箱”里的东西
13
93
4
410
48,233
Cheng1937(∞) retweeted
MCP - 让 Agent 调用工具有了标准 A2A - 让 Agents 之间对话有了标准 今天 Google Deepmind 发布了 Agent2Agent Protocol (A2A) 以及一套开源的 Agent Development Kit (ADK),不到 100 行代码就能构建一个完整 Agent 还支持 MCP!
14
114
2
528
63,926
Cheng1937(∞) retweeted
谷歌昨晚最大的发布是这个 AI 编程工具 firebase studio 这玩意太猛了,充分利用谷歌的基建 可以看作是 lovable+cursor+replit+windsurf 合体应用 支持一键构建后端、前端和移动应用 从生产到发布都在一个地方完成,除了网页还有安卓应用 应用发布后还带有数据监控能力
Introducing @Firebase Studio, a new cloud-based, agentic development environment with everything developers need to create full-stack AI apps quickly, all in one place. Start building now → goo.gle/4cvcUzY #GoogleCloudNext
17
161
7
794
174,870
Cheng1937(∞) retweeted
一个可帮你在家快速搭建 AI 集群的开源项目:exo。 该项目支持将现有设备统一到一个功能强大的GPU中,支持 iPhone,iPad,Android,Mac,Nvidia,树莓派等等几乎所有设备。 GitHub:github.com/exo-explore/exo 在项目部署成功后,一旦有新设备联网,即可接入使用,充当 AI 算力集群的一部分。
7
70
2
279
22,831
Cheng1937(∞) retweeted
来自博主 Akshay 分享的一套构建自主纠正的 RAG 工作流。 除了能够搜索我们的文档外,还能在有需要时进行网络搜索,进一步确保内容准确性。 此外,跟大家分享一下制作这样的动态图,可以使用一个开源工具:draw.io。 教程:drawio.com/doc/faq/connector… GitHub:github.com/jgraph/drawio-des…
5
51
2
235
26,594
Cheng1937(∞) retweeted
有关 AI 编程,我觉得近期最值得看的一篇文章和视频都来自 Addy Osmani 《AI 辅助编码的残酷真相:它能帮你完成70%的工作,但最后30%令人非常沮丧》 今天终于抽空把视频给翻译了一下,见评论
16
131
7
646
113,083