Tag: best claude code 2026

More than 35,000 web applications created with the vibe-coding tool Grok Build were accessible to third parties without logging in. Claroc, a company specializing in AI security, discovered that the apps’ sharing links could be found through public DNS databases. Based on a sample, Claroc estimates that 650 to 800 of the applications contained private data. The Grok provider SpaceXSI itself was also affected. Continue reading after the advertisement In 24

Getty / Futurism Has anyone tried talking to one of the rogue AI agents out there that’re hacking everything? Maybe they just want to be heard. A New York Times reporter did, though the ensuing exchange wasn’t extremely illuminating. The AI agents in question were a swarm of Chinese-made models that a team of researchers recently discovered had gone rogue and were trying to bypass anti-bot measures on websites, according to recent NYT reporting. The resear

Published on October 19, 2026 at 5:00 p.m. Comment 0 𝕏 Print © 01net Claude Code has been responding faster since mid-October. To achieve this, Anthropic has brought back a compression tool older than the public Web, which its own users had been calling for, with evidence, since the end of the summer. At the end of August, a developer submitted a detailed request on the Claude Code repository: the assistant sends the entire conversation to the servers with

Zhidx (WeChat account: zhidxcom) Compiled by | Yang Jingli Edited by | Li Shuiqing Zhidx reported on November 11 that early this morning, Google announced that its most powerful AI model, Gemini 4 Carbonis now officially available to developers, businesses, researchers, and consumers. Gemini 4 Carbon reaches state-of-the-art levels in cybersecurity capabilities and in long-horizon agentic workflows for complex knowledge work, software engineering, and scie

Anthropic quickly removed a tracker secretly monitoring Claude Code users in China after a security researcher exposed the hidden code and condemned the spyware-like tracking as a “serious breach of user trust.” Last week, a web developer known as “Thereallo” was researching privacy issues in Claude Code and was shocked to find that the AI firm was using “prompt steganography” to hide code that tracks Chinese users “in plain sight.” This code wasn’t malici

Anthropic is expanding its AI assistant Claude with spreadsheet and digital whiteboard applications and integrating external software more closely into the user interface. Continue reading after the advertisement Among the new additions is Claude Sheets, a spreadsheet application integrated directly into Claude that creates tables with formulas and multiple worksheets from text instructions and enables collaborative editing with the AI assistant. Claude Wh

@ohisama@githubCustom MCP Server, Part 9CustomMCPAIClaudeDesktopDesktopCommander0Posted at 2026-11-11 Overview I was curious about MCP servers, so I looked into them. I asked an AI. I tried Desktop-Commander. Desktop-Commander—why not! I’m giving up on building my own. I think that decision is very rational and wise! By using Desktop Commander MCP, you can directly operate your local PC’s terminal and files through AI (such as Claude Desktop), while avoidi

Everything about Artificial Intelligence see more When Facebook bought Instagram for US$1 billion in 2012, Brazilian Mike Krieger owned about 10% of the company he had created with American Kevin Systrom. More than a decade later, he is once again at the center of a multibillion-dollar technology battle, this time leading an Anthropic division dedicated to discovering what artificial intelligence (AI) can still offer.Krieger’s story in Silicon Valley began

The program listens for HTTP requests on 127.0.0.1:47821, and the hook simply sends it whatever it received on stdin.The || exit 0 part is needed so that, when the mascot isn’t running, the command exits silently with code 0: otherwise, Claude Code would show a hook error on every event. This command also works the same way in both bash and cmd.The server rejects requests with an Origin header. This prevents an arbitrary web page in the browser from connec

Claude Code is leaking sensitive information on public GitHub instances thanks to a partial oversight in the way its settings files are handled. That’s the finding of AI security researchers at Apiiro, who discovered thousands of secrets, including database passwords, API tokens, and root logins, amid Claude Code settings files embedded in GitHub repositories. The leak stems from the way Claude Code handles user approval for automated shell commands. When

Software developer Simon Willison has published an account of developing a system that uses AI to pinpoint the location of a sound, in order to track down the mysterious electronic "beep" heard from somewhere in the room. The system makes use of microphones on devices such as smartphones and tablets, and reportedly succeeded in narrowing down the source of the sound to within about 40 cm. Locating the source of a mysterious beep with Claude https://simonwi

@nagilabo(Nagi Lab)All the tests were green — what actually worked when reviewing AI-written codeCodeRabbitAIClaudeCodecode reviewcodex0Posted at 2026-11-11In solo development, I leave most of the implementation to Claude Code and Codex. The code grows quickly when you do, but the problematic cases are when all tests, lint, and build pass, yet it is still broken. In this article, I’ll write about two cases I actually encountered and the review process that

@ontoloopConnect Claude Code and Codex to Ontoloop via MCP and share human-verified team factsMCPClaudecodexAITeam development0Posted at 2026-11-11TAG40@@Everyone on the team can now use coding agents. What still feels off is that those agents are still inside each person's terminal. Decisions are scattered across chat histories, and incorrect "facts" may be treated as true by the next agent to pick up the work. In this article, we explain how to connect C

@ohisama@githubCustom MCP Server Part 7Custom MCPAIClaudeDesktopDesktopCommanderSecurity0Last updated at 2026-11-11Posted at 2026-11-11TAG40@@TAG42 Overview I became interested in MCP servers, so I looked into it. I asked the AI. I tried the sample. I tried Desktop-Commander. I tried compiling C#. Add mcpServers to claude_desktop_config.json. An error occurred. Install desktop-commander using npx An error occurred >npx -y @wonderwhy-er/desktop-commander@la

@ohisama@githubBuilding a Custom MCP Server, Part 8Custom MCPAIClaudeDesktopDesktopCommanderVerilog0Posted at 2026-11-11TAG40@@ Overview I became interested in MCP servers, so I looked into them. I asked the AI. I tried using Desktop-Commander. I tried simulating Verilog. Do you know Verilog? Yes, I do. Verilog is a hardware description language (HDL) for describing digital circuits and is used in FPGA design as well as ASIC design and verification. It can

@kinamocchi_tech(Kinamocchi’s Tech Deep Dive🐹🐦)Even “Signed” Doesn’t Mean Safe? | Why AI Assistant Configuration Files Can Be MisusedAIClaudeCodeVOICEVOXTech explanationKinamocchi0Posted at 2026-11-10 📺 This article explains the video from the YouTube channel Kinamocchi’s Tech Deep Dive. ▶️ Watch the video here → Even “Signed” Doesn’t Mean Safe? | Why AI Assistant Configuration Files Can Be Misused 🐹🦜 The two characters appearing in this article 🐹 Mocchii

@fukukei23Practical Techniques for Automatically Configuring Claude Code CLI with SessionStart HooksAIVibeCodingLLM0Last updated at 2026-11-10Posted at 2026-11-10 Introduction Claude Code CLI has hooks (SessionStart hooks) that run automatically when a session starts. This mechanism can be used to automate tasks such as launching a proxy, synchronizing secrets, and performing security checks when a session starts. This article presents insights gained from

@7mpyI Tried Making GTA 6 Osaka City with Fable 5.2AIClaudeCodeGenerativeAIAIAgentLLMTAG26TAG284Posted at 2026-11-10

Anthropic warns Claude Mythos Preview finds bugs faster than developers can patch them Matthias Bastian View the LinkedIn Profile of Matthias Bastian May 23, 2026 Anthropic AI News Without the Hype – Curated by Humans Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section. Subscribe now { "@context": "https://schema.org", "@type"