Hi all, here are today's three links:

Cowork: Claude Code for the rest of your work

Anthropic releases Claude Code adapted for non-technical users:

In Cowork, Claude completes work like this with much more agency than you’d see in a regular conversation. Once you’ve set it a task, Claude will make a plan and steadily complete it, while looping you in on what it’s up to.

How to build agents with filesystems and bash

Vercel points out that sometimes files on disk is all you need:

If agents excel at filesystem operations for code, they'll excel at filesystem operations for anything. Agents already understand filesystems.

Recursive Language Models

Prime Intellect offers a new solution for improving agent performance with long contexts:

We believe that teaching models to manage their own context end-to-end through reinforcement learning will be the next major breakthrough, enabling agents to solve long-horizon tasks spanning weeks to months.

If you have any agent-related links you'd like to feature, let me know!

Cheers,
Ivan