datasette-agent 0.2a0 Released: Tools Can Ask Users Questions During Execution
Simon Willison's Weblog3 days agoReleasedatasette-agent 0.2a0 lets tools ask users questions during execution through ToolContext. Unanswered questions suspend the agent turn, render as chat UI forms, and persist across server restarts. A new save_query tool can store agent-written SQL as a Datasette saved query, but only after explicit human approval.
Apple's AI Can Now Change Your Passwords. What Could Possibly Go Wrong?
Hacker News (AI keywords)5 days agoCommentaryApple's AI assistant has gained the ability to change account passwords on behalf of users, raising eyebrows in the security community. The author uses pointed sarcasm to question whether delegating password management to an AI system is wise. This development reflects a broader trend of AI agents gaining deeper OS-level permissions, blurring the line between helpful automation and dangerous over-trust.
Connect the dots: Build with built-in and custom MCPs in Studio
Mistral AI News6 days agoNew ToolMistral AI released Connectors in Studio as a public preview for grounding AI apps in enterprise data. Developers can register reusable built-in or custom MCP connectors and use them through APIs, SDKs, conversations, completions, and agents. The release adds direct tool calling, connector governance, tool availability controls, and human-in-the-loop approval before sensitive tool execution.
Cognition’s Scott Wu says AI coding agents shouldn’t replace humans
TechCrunch AI16 days agoCommentaryCognition makes Devin, described by TechCrunch as the first and arguably most successful AI coding agent. Scott Wu says the product is not meant to supplant human programmers. The key takeaway is a positioning statement: AI coding agents are being framed as tools for software work, not as a direct removal of humans from development.
Show HN: Continue? Y/N, a 60-Second Game About AI Agent Permission Fatigue
Hacker News (AI keywords)17 days agoCommentaryThis Show HN submission points to “Continue? Y/N,” a 60-second game about AI agent permission fatigue. With no article body provided, the available information suggests an interactive commentary on how repeated approval prompts can wear users down. The project appears most relevant to developers, designers, and product teams thinking about agent UX, consent flows, and trust boundaries.
透過 AI Agent、沙盒與人工協作，讓 Turborepo 速度提升 96%★ 75
Vercel Changelog76 days agoReleaseVercel recently shared a highly instructive technical case study, demonstrating how they leveraged AI Agents, secure Sandboxes, and human engineer…
AssistLoop 正式加入 Vercel Agents 市集，為 AI Agent 引入人機協同機制
Vercel Changelog135 days agoReleaseVercel officially announced that AssistLoop, an AI collaboration platform focused on "Human-in-the-Loop (HITL)" mechanisms, has joined the Vercel Agents…
如何使用對抗性數據動態訓練你的模型 (以 MNIST 為例)
Hugging Face Blog1,429 days agoTutorialThis technical blog post from Hugging Face takes an in-depth look at how to build a closed-loop system for "Dynamic Adversarial Training." In a traditional…