Bleeping Computer

Backdoored Telnyx PyPI package pushes malware hidden in WAV audio

TeamPCP hackers compromised the Telnyx package on the Python Package Index today, uploading malicious versions that deliver credential-stealing malware hidden inside a WAV file. Earlier today, the ...

Supply chain attack on LiteLLM: Affected parties should change credentials

An attack on the open-source library for connecting to LLMs has apparently occurred, allowing two compromised packages to ...
腾讯网

Karpathy紧急警告：月下载近亿的LiteLLM被投毒！已有开发者因Cursor MCP ...

但是也有人质疑卡帕西的“利用LLM提取功能”的这一措施，表示“只是把一个未经审查的代码库换成了一个LLM输出的而已”。这个就比较见仁见智了，使用LLM过滤一遍对提高代码安全性是否存在帮助依然非常依赖提示词。
腾讯网

学习笔记：从 Agent 到 Skills — AI 智能体架构的范式转变

报告日期：2026-02-28 关键词： Agent Skills, MCP, OpenClaw, A2A, Agentic AI, 模块化架构一、谁提出了从 Agent 到 Skills 的转变？1.1 起源：Anthropic 的两步棋Anthropic 在不到 14 个月内连续发布了两个开放标准：Anthropic 工程博客原文："Building a skill for an agent ...