今年3月，腾讯发布了《2026年AI人才报告》，其中提到“AI辅助编程工具使通用型开发任务效率提升约50%”。这个数字在测试圈的讨论群里引发了一轮激烈争论。不是因为50%有多吓人，而是因为测试本身就是一道“执行质量”的防线——如果连执行者都在被加速，这道防线还能守多久？ 更具体的信号在最近两个月开始密集出现。 4月份，JetBrains Rider 2026.2 EAP发布了一个新特性：AI ag ...

Claude now authors 80% of Anthropic's production code. The company's new paper maps the path to recursive self-improvement and calls for a global pause mechanism.

VS Code needs to catch up.

Discover pro-level techniques for Claude Code, including memory management, Auto Mode automation, and context architecture to ...

I don't waste time correcting Claude Code anymore.

Anthropic has published an unusually concrete account of how its own engineers use Skills in Claude Code, the company’s command-line coding agent, in a June 3 post on the Claude blog written by ...

Combining Claude Code with Obsidian creates a structured yet adaptable system for managing information and projects. In a recent breakdown, Chase AI explores how this integration uses Obsidian’s ...

A Claude Code GitHub Action flaw let one malicious issue hijack repositories via prompt injection. Anthropic has patched it.

A flaw in Anthropic’s Claude Code GitHub Action allowed a malicious GitHub issue from a bot actor to trigger workflows and ...

A flaw in Claude Code's GitHub Action let attackers bypass permission checks via fake bots and steal OIDC tokens through prompt injection.