Tau Will Ask Before It Reads a Project’s AGENTS.md
Tau’s new Project Trust layer turns local instructions, skills, prompts, themes, and extension candidates into a directory-specific decision before they become ambient agent input.
More From This Edition
All editions ›Recent Editions
Full history ›Qwen Code Stops Calling a Rerun a Resume
Two September 18 changes separate a recoverable interruption from a fresh start: refuse a missing journal, then make the interrupted run visible. Neither promises automatic recovery.
Unsloth Rolled Back the Sandbox Its Review Had Just Cleared
A live review reported working Linux and macOS isolation. The next day, the implementation was reverted. Studio’s remaining protections make the consequence more specific than “the sandbox is gone.”
The Allow Button Was Waiting for a Socket
OpenWork reproduced a ten-second approval timeout before the user’s decision reached the engine. The repair gives that decision a separate route through a busy desktop app.
The Cookie Jar Between Two Agent Sessions
Microsoft’s Agent Framework is changing what its Python HTTP clients remember. A new conversation can still inherit an old login when both use the same cookie jar.
How AI Agents Work
The long-form map behind the daily paper: context, tools, loops, memory, delegation, safety, interface, and trust.