Hackers breached the online platform of the open-source AI coding agent OpenCode and stole some users' API keys.
The article explains a tmux setup for efficiently running AI agents like Claude Code, covering session naming, pane layouts, scripting via send-keys, and the limits encountered when running four agents simultaneously.
The development background, usage, and benchmarks of Superlite, an autonomous build workflow for Claude Code that does not ask for permission every four minutes, are introduced.
A security breach linked to a large-scale distillation attack leaked credentials for 1.4% of OpenCode Console users.
The Claude Code team optimizes for the minimum amount of human review time that does not cause reverts or incidents, addressing the volume of code produced by agents.
Microsoft researchers published a paper analyzing the usage and productivity impact of command-line AI coding tools among tens of thousands of internal engineers.