So, every time I make a new project, do I need to create a New Directory Folder, or just /init everytime i want a new one in the same directory? And how will I know it doesnt messed with the existing one I created?

If you have a Claude.md file do you still Need to have an agents.md file as well as a GSD? Thank you

If I Connect through MCP in Claude code or cowork, does the connection work in both tools or just in the one i use to Connect it?

Does anyone here have experience rolling out Claude Code across a team? I’m thinking of two setups: - DevOps: ~3–5 people - Engineering: ~12–18 people I’ve got a couple ideas, but the core problem I’m trying to solve is: how do you enforce guardrails + consistent patterns so people use Claude Code safely and the same way. In most cases, PRs are the real gatekeeper for catching damaging changes. Even with tools like Greptile or CodeRabbit, we still require human review and that’s not changing (SOC2). But here’s the sharp edge: a few of us are Git admins, and that’s where things can get nasty. Admins can override protections and bypass merge requirements. I’ve had this happen twice: I was running in agent mode and Cursor performed a git commit and effectively bypassed our merge requirements by committing directly. After the first time, I set a hard rule in Cursor globally: no git commits, ever. That worked… for a while. Then it happened again anyway. When I asked Cursor why it ignored the rule, the response was basically: “Sorry, you’re right, there is a hard rule — I’ll try to do better next time.” I’m not even exaggerating. So yeah — how are you all handling: - org/team-level guardrails for Claude Code/Cursor agent behavior - preventing “agent commits” (especially for admins) - standard patterns for safe workflows across teams Im curious what other people have seen.

Hey, I'm wondering if anyone of you has experience in making AI content? I would need like a 30-60 second clip "interview style" Appreciate any insights!