Spec-driven development: Using Markdown as a programming language when building with AI
I coded my latest app entirely in Markdown and let GitHub Copilot compile it into Go. This resulted in cleaner specs, faster iteration, and no more context loss. ✨

The post [Spec-driven development: Using Markdown as a programming language when building with AI](https://github.blog/ai-and-ml/generative-ai/spec-driven-development-using-markdown-as-a-p … ⌘ Read more

AI & LLM Addiction In School ⌘ Read more

Docker MCP Toolkit: MCP Servers That Just Work
Today, we want to highlight Docker MCP Toolkit, a free feature in Docker Desktop that gives you access to more than 200 MCP servers. It’s the easiest and most secure way to run MCP servers locally for your AI agents and workflows. The MCP toolkit allows you to isolate MCP servers in containers, securely configure… ⌘ Read more

AI Therapy & The Risks of Cheap Mental Health ⌘ Read more

AI’s Impact on Intimate Relationships ⌘ Read more

MSI EdgeXpert Compact AI Supercomputer Based on NVIDIA DGX Spark
The MSI EdgeXpert is a compact AI supercomputer based on the NVIDIA DGX Spark platform and Grace Blackwell architecture. It combines a 20-core Arm CPU with NVIDIA’s Blackwell GPU to deliver high compute density in a 1.19-liter form factor, targeting developers, researchers, and enterprises running local AI workloads, prototyping, and inference. The EdgeXpert achieves up […] ⌘ Read more

The Trust Paradox: When Your AI Gets Catfished
The fundamental challenge with MCP-enabled attacks isn’t technical sophistication. It’s that hackers have figured out how to catfish your AI. These attacks work because they exploit the same trust relationships that make your development team actually functional. When your designers expect Figma files from agencies they’ve worked with for years, when your DevOps folks trust… ⌘ Read more

Run, Test, and Evaluate Models and MCP Locally with Docker + Promptfoo
Promptfoo is an open-source CLI and library for evaluating LLM apps. Docker Model Runner makes it easy to manage, run, and deploy AI models using Docker. The Docker MCP Toolkit is a local gateway that lets you set up, manage, and run containerized MCP servers and connect them to AI agents.  Together, these tools let… ⌘ Read more

Using AI to map hope for refugees with UNHCR, the UN Refugee Agency
With the help of GitHub, UNHCR turned drone imagery into maps — helping refugees in Kakuma and Kalobeyei build sustainable, powered communities.

The post Using AI to map hope for refugees with UNHCR, the UN Refugee Agency appeared first on [The GitHub Blog](https://github. … ⌘ Read more

A step-by-step guide to modernizing Java projects with GitHub Copilot agent mode
Learn how to use GitHub Copilot agent mode to modernize legacy Java projects with guided upgrades, automated fixes, and cloud-ready migrations.

The post A step-by-step guide to modernizing Java projects with GitHub Copilot agent mode … ⌘ Read more

MCP Horror Stories: The Drive-By Localhost Breach
This is Part 4 of our MCP Horror Stories series, where we examine real-world security incidents that expose the devastating vulnerabilities in AI infrastructure and demonstrate how Docker MCP Gateway provides enterprise-grade protection against sophisticated attack vectors. The Model Context Protocol (MCP) has transformed how developers integrate AI agents with their development environments. Tools like… ⌘ Read more

Gartner positions GitHub as a Leader in the 2025 Magic Quadrant for AI Code Assistants for the second year in a row
Our commitment is to empower every developer and stay true to our north star by building an open, secure, and AI-powered platform that defines the future of software development.

The post [Gartner positions GitHub as a Leader in the 2025 Magic Quadrant for AI Code Assistants for the second yea … ⌘ Read more

Framework Desktop -

AMD Ryzen™ AI Max 385 - 32GB
32GB LPDDR5x-8000 memory (soldered)

AMD Ryzen™ AI Max+ 395 - 64GB
64GB LPDDR5x-8000 memory (soldered)

AMD Ryzen™ AI Max+ 395 - 128GB
128GB LPDDR5x-8000 memory (soldered)

Beyond Containers: llama.cpp Now Pulls GGUF Models Directly from Docker Hub
The world of local AI is moving at an incredible pace, and at the heart of this revolution is llama.cpp—the powerhouse C++ inference engine that brings Large Language Models (LLMs) to everyday hardware (and it’s also the inference engine that powers Docker Model Runner). Developers love llama.cpp for its performance and simplicity. And we at… ⌘ Read more

Does Apple really need to become an AI company?
Amid claims it is falling behind the likes of Google and OpenAI, the iPhone-maker continues to double down on what it’s actually good at. ⌘ Read more

From hallucinations to High Court: Can AI administer justice?
There are calls for Australian courts to consider trialling AI technology, and even allow it to decide cases, despite a growing number of fabrications and errors. ⌘ Read more

https://jacobin.com/2025/09/universities-corporatization-neoliberalism-ai-outsourcing

Since Google announced their intentions to heavily limit sideloading on Android, starting end of 2026, I’ve been looking for potential solutions, for this policy change, that threatens the majority of projects I maintain, in some way. Google already killed my browser project years ago, but I have no other choice, than to fight this, any way I can.

The best choice to deal with this, will probably be the Android Debug Bridge, which can be used not only to install apps unrestricted, but also to uninstall, or remove, almost any unnecessary part of the OS. Shizuku, combined with Canta Debloater, is the winning combination for now.

I’ve already removed most Google apps from my device: the annoying AI assistant, the stupid Google app adding the annoying articles, left of your homes screen, Google One, Gboard, Safety app… it’s amazing, no distracting Google slopware, like in the good old Android 2 days! And I absolutely intend to keep it this way, from now on, no new Google apps or services on my devices, unless Google can give me a good enough reason, to allow them there and whenever the app that verifies signatures, to block installing apps not approved by Google, I’ll just remove it from my device and advocate others do so too.

How about no longer using in-browser Git repo viewers? Make the AI bots do the work and actually clone the repo.

@movq@www.uninformativ.de Interesting, yes. I didn’t know that.

No AI being used is really great. However, the same clips shown over and over again and some images being mirrored was quite annoying to me. Also, there were some quite terrible computer animations and sometimes the narration and picture didn’t match at all. Talking about the medieval period and then showing an image from the 18th hundred or so. What the heck?

These production issues made me sceptical pretty much early on. So I quickly crosschecked Wikipedia. But it seems spot on from what I’ve read. Very good. Also, the narrator’s voice was really nice to listen to.

Eels are fascinating creatures. :-)

How Generative AI Video Works - Computerphile ⌘ Read more

Hmm, gnu.org is slow as heck. Shorter HTML pages load in about ten seconds. This complete AWK manual all in one large HTML page took a full minute: https://www.gnu.org/software/gawk/manual/gawk.html Is there maybe some anti AI shenanigans going on?

In any case, I find the user guide super interesting. My AWK skills are basically non-existent, so I finally decided to change that. This document is incredibly well written and makes it really fun to keep reading and learning. I’m very impressed. So far, I made it to section 1.6, happy to continue.

@prologic@twtxt.net I’m doing that now as well, but I don’t think this is a good solution. This is going to hurt “self-hosting” in the long run: I cannot afford true self-hosting where I actually do host everything here at home – instead, I must use a cloud provider / VPS for that. It is only a matter of time until my provider starts doing AI shit as well (or rather, the customers do it) and then what? I get blocked, e.g. I can’t send email to (some) people anymore. This is already bad and it’s going to get worse.

Hahaha, how funny is that!? The Dunning-Kruger effect research was sparked off by two bank robbers who rubbed lemon juice in their faces as this makes them invisible, just like invisible ink. :‘-D https://en.wikipedia.org/wiki/1995_Greater_Pittsburgh_bank_robberies

I now imagine all these AI wankers with lemon juice in their faces.

“But all your stuff is MIT licensed! They are allowed to do that!”

Haha. As if they would care. They crawl everything they get their hands on.

Besides, that’s not true, the license states that the copyright notice must be retained. “AI” breaks that. They incorporate my code and my articles in their product and make it appear as if it was their work.

Why do I care about this?

1. The load will become a problem at some point.
   
2. These crawlers and the current “AI” in general are breaking the rules. I am supposed to be paying for every little thing, I get sued for “piracy”. But apparently, these rules only apply to me. If I had more money, I could break them. Fuck that.
   
3. I simply don’t want it. Period.

@thecanine@twtxt.net We don’t use Microsoft at work – but similar products of other big companies. They’re all doing the same. The core product gets worse and worse, because they focus so much on vomiting “AI” over everything.

It will die down eventually. I hope.

We use all the Microsoft programs at work - Teams and Outlook especially.

After all kinds of technical problems with Teams, that sometimes go unresolved for over a year, Microsoft shifted their priorities away from fixing things and towards adding an annoying AI Copilot button, that just takes up space and all it does, is loads the website in Teams, so I disabled it. Soon they just add it back, but in a different row of icons, therefore it’s now a different button, you have to disable (I think they added yet another one, to the Teams, on my work phone and I had to disabled that too). Not too long after, the desktop one just enabled itself, because of “an error” and I can disable it, but doing so activates a popup, that begs you to turn it back on, every once in a while. You can’t disable the popup and can only click “Yes” or “Not now” on it. I still keep it disabled, out of principle, but yesterday I noticed yet another Copilot button, this time in the top right corner of my Outlook and this one cannot be disabled, on the business version of Outlook and even on the personal one, it’s only possible to do it through hidden privacy settings, by prohibiting the program from connecting to Microsoft servers, for extra “features”.

There’s people complaining about it online, so it’s clear nobody really wants it, but at this point Microsofts position is that you will have at least one useless AI button on your screen, at any given time, and you will be happy. And yes, their AI sucks and if I absolutely have to use AI for something, there’s already 2 better options, we have access to, at work.

Erlang Solutions: Healthcare Blog Round-Up
Healthcare is moving quickly, and technology is playing a big part in that shift. The way information is collected, the way patients are cared for, and the way hospitals run are all changing.

Over the past year, our team has written about some of the most important trends shaping the future of healthcare. In this round-up, we bring together three of those articles: remote patient monitoring, big data, and generative AI.

Maybe you have been following along, or … ⌘ Read more

Ditt jobb kommer ersättas av AI ⌘ Read more

Another wave of tens of thousands of hints by the same bot on the same file:

There’s probably a simple explanation for this: Maybe this bot was written with “AI” and it’s simply complete garbage.

This isn’t a serious threat for my low-profile website – yet. Can’t wait for this to get worse …

Kunne det være sandt, at en persons foragt for LLM’er er proportinel med mængden af AI-markedsføring som personen er blevet vist? ⌘ Read more

Comment on Newelle, AI “Assistant” for GNOME, Hits Version 1.0 by temp mail
Ne zamandır web sitelerim için aradığım içeriği sonunda buldum. Bu kadar detaylı ve net açıklama için teşekkürler. ⌘ Read more

Bloody AI clowns:

How many IPs am I supposed to block, eh?

This: https://www.thebignewsletter.com/p/why-are-we-pretending-ai-is-going

Nuke it from orbit: https://www.aaron.ai/

Nooooooooooooooooooooooooooooooooooooooooooooooooooooooo, the doctors have started using AI voice agents and they understand jack shit. 😭😭😭

萬字解讀：8 種常見框架，選擇哪一種來開發 MCP 呢？
模型上下文協議 (Model Context Protocol，MCP) 是一個新標準，用於以統一的方式將 AI 助手 (如 llm) 與外部數據源和工具連接起來。自從 MCP 引入以來，出現了各種各樣的框架來幫助開發人員更容易地構建 MCP 服務器。關於 MCP 的更多介紹，可以參考拙作——在本文中，嘗試評估 8 種常見的 MCP 服務器開發框架，每種框架的語言或生態系統存在不同，並對其易用性、 ⌘ Read more

一文搞懂大模型的數據集 FineWeb：讓 AI 更聰明的 15 萬億字數據集
你有沒有想過，爲什麼有些 AI 回答問題時邏輯清晰、知識淵博，而有些卻答非所問、胡說八道？關鍵就在於它們 “喫” 了什麼樣的數據。就像人類的成長需要優質教育一樣，AI 的訓練也離不開高質量的數據。但在 AI 領域，一直存在一個尷尬的現狀：那些表現最好的 AI 模型，比如 GPT-4、Claude 等，它們的訓練數據都是商業機密，普通研究者和小公司根本接觸不到。這就像最好的學校不對外開放，只有少數人 ⌘ Read more

MCP 規範完整中譯稿：2025-3-26 版
【引】儘管 AI 可以幫助我們順利地理解 MCP 規範，但一份完整的 MCP 規範中譯稿還是有意義的，可以進一步幫助我們理解 MCP 規範的來龍去脈，以及協議中細節的方方面面。如果希望希望極簡入門的話， 可以閱讀老碼農的新作——1. 規範模型上下文協議 (Model Context Protocol，MCP) 是一個開放的協議，支持 LLM 應用程序與外部數據源和工具之間的無縫集成。無論是構建基於 ⌘ Read more

LangGraph 多智能體羣：使用 LangGraph 創建羣風格多智能體系統的 Python 庫
LangGraph 多智能體羣是一個 Python 庫，旨在將多個 AI 智能體編排成一個有凝聚力的 “羣”。它建立在 LangGraph 之上，LangGraph 是一個用於構建健壯、有狀態智能體工作流的框架，以實現一種特殊形式的多智能體架構。在一個羣中，具有不同專業的智能體根據任務的需要動態地將控制權交給彼此，而不是單個單一智能體嘗試處理所有事情。系統會跟蹤最後一個激活的智能體，以便當用戶提供 ⌘ Read more

AI this, AI that.

Tech is no longer interesting. I need to find a new field.

FeatherScan v4-0 - 一款 Linux 內網全自動信息收集工具
前言–在平時滲透打靶的時候，經常要自己手工輸入命令，做各種基本的信息收集，非常的繁瑣，所以自研了一款工具，這款工具沒有接入 AI，因爲不合適，接入了 AI 的話在一些不能上網的環境下進行信息收集，權限提升的分析，會非常的不方便，這款工具全都在目標機器本地執行（執行速度快，提高滲透測試效率），類似於 fscan，需要上傳到目標靶機上，後期會增加離線的 POC 和漏洞庫對 linux 系統進行全面的 ⌘ Read more

深度解密 A2A 協議：開啓智能體協作的新紀元
引言：智能體時代的互操作性挑戰隨着人工智能技術的飛速發展，AI 智能體（AI Agents） 正成爲構建複雜 AI 應用的新範式。它們不再僅僅是簡單的問答機器人，而是具備感知、推理、決策、行動能力的獨立 “個體”。然而，當這些智能體由不同的團隊開發，運行在不同的平臺，甚至使用不同的技術棧時，一個核心挑戰便浮出水面：它們如何才能像人類團隊一樣，順暢地相互理解、溝通並高效協作？想象一下，一個能夠進行市 ⌘ Read more

智能體間協作的 “巴別塔困境” 如何破解？解讀 Agent 通信 4 大協議：MCP-ACP-A2A-ANP
AI 智能體的興起觸發了 AI 應用協作的新領域。這些智能體不再侷限於被動的聊天機器人或獨立的系統，它們現在被設計用於推理、計劃和協作ーー跨任務、跨域甚至跨組織。但隨着這一願景成爲現實，一個挑戰很快浮出水面： 智能體如何以一種安全、可伸縮和可互操作的方式可靠地相互交流、共享上下文並共同做出決策？一類新的通信協議應運而生。從模型上下文協議 (MCP) 到 IBM 和思科的智能體通信協議 (ACP) ⌘ Read more

go-nanoid：Go 短 ID 生成庫
“An amazing level of senseless perfectionism, which is simply impossible not to respect.“go-nanoid (github.com/matoous/go-nanoid) 是 JavaScript 知名庫 ai/nanoid (github.com/ai/nanoid) 的 Go 語言實現版本。原版 nanoi ⌘ Read more

go-nanoid：Go 短 ID 生成庫
“An amazing level of senseless perfectionism, which is simply impossible not to respect.“go-nanoid (github.com/matoous/go-nanoid) 是 JavaScript 知名庫 ai/nanoid (github.com/ai/nanoid) 的 Go 語言實現版本。原版 nanoi ⌘ Read more

基於 Dify 的 RAG 知識庫搭建
Dify 是一款開源的大模型應用開發平臺，旨在幫助開發者快速構建生產級生成式 AI 應用。在 Dify 本地化部署中，知識庫功能是實現企業級 AI 應用的核心能力。本文介紹基於版本 1.5.1 搭建知識庫全流程解析，包括以下內容：Dify 基本概念Dify 本地部署基於 Dify 的知識庫搭建一、Dify 基本概念Dify 是一款開源的大模型應用開發平臺，旨在幫助開發者快速構建生產級生成式 AI ⌘ Read more

一文帶你讀懂 Google LangGraph 項目，快速入門 AI Agent 全棧開發
一、項目背景與目標———最近在帶着同事一起做智能 Agent 相關的內部項目，發現很多人對 LangGraph 非常感興趣，但又不太清楚如何從零開始搭建一個完整的 AI Agent，我於是在 github 上找，看看有沒有好的開源項目給他們學習，偶然間發現了 google-gemini 開源的這個項目 [1]，正好拿來給他們講講，學習學習。發現整理的材料又正好可以出一期公衆號文章，就作 ⌘ Read more

停止構建 AI Agent！這裏有 5 個更簡單的 LLM 工作流模式，能解決 90- 的問題
大家好，我是 Tony Bai。如果你正在開發 AI 應用，你很可能聽說過、嘗試過，甚至正在掙扎於構建一個 “AI Agent”。我們都看過那些令人心潮澎湃的 Demo：一個 AI Agent 被賦予一個目標，然後它就能自主地規劃、調用工具、瀏覽網頁、編寫代碼，最終完成任務。於是，我們紛紛投身其中，搭建記憶系統、定義工具、編寫角色背景…… 感覺就像在創造一個真正的數字生命，充滿了力量和進步感。但現 ⌘ Read more