A majority of Claude interruptions are asking whether it's okay to run some command. This is why --dangerously-skip-permissions is so powerful: we can achieve so much more per unit prompt.

But why is that the choice? Why must we choose between becoming a pedantic babysitter or giving Claude's actions dangerous free reign over our system?

Clash gives us the best of both worlds. It allows sandboxing what matters: the actions Claude is performing. Every tool, every command, independently.

The Clash plugin for Claude runs in two phases,

1. matching and deciding what actions Claude can perform, and
2. dynamically selecting the sandbox in which that action will be performed.

Unlike wrapping our agent in a container or VM, we are in control of which tool gets what access.

With Clash, we get the productivity of --dangerously-skip-permissions without losing the control.

Software engineering is a team sport, but Claude can make the team feel slow.

For those of us who've jumped down the rabbit hole of agentic engineering, there's can be a feeling of flying. An unbridled joy of saying what we want and seeing it happen before us, but this often hits a hard reality when we coordinate with others.

We sit in our accelerated silos and the bandwidth of communication between them can't keep up. Either we slow down to work at the speed of yesterday's tooling, or we leave behind decades of engineering practice, bushwacking into the uncertain future of possibility.

With Workshop, we're solving that. Zero friction collaboration-first development built on a foundation of this brave new world.

Start by using Claude just the way you always have. Invite a teammate to join you. Use the rich UI to easily search past conversations and leverage them for durable workflow improvement.

Software is a team sport.

Our agents should augment our team, not divide us.