For what it's worth, I'm using plain Claude Code: one git worktree per session, the sessions messaging each other directly, and one coordinator session that assigns the work. To be honest, it fills the multi-agent gap…
[flagged]
If you had been using Claude Code's built-in sandbox, would it have stopped this from happening? I'm not using it myself and I'm new to this, so I'm trying to understand. Thanks for sharing btw
For what it's worth, I'm using plain Claude Code: one git worktree per session, the sessions messaging each other directly, and one coordinator session that assigns the work. To be honest, it fills the multi-agent gap…
[flagged]
If you had been using Claude Code's built-in sandbox, would it have stopped this from happening? I'm not using it myself and I'm new to this, so I'm trying to understand. Thanks for sharing btw