A free guide by New Money School
Pro Tips
You are probably giving Claude more permission than you realize. Before you connect it to Gmail, Google Drive, or Slack, do this one thing first. This is part 4 of my AI safety series.
Most people connect Claude to their tools and never check what they just handed over. I did the exact same thing at first.
When you connect Claude to Gmail, Google Drive, or Slack, you are giving it a door into real data. Your email. Your files. Your team messages. That is a big deal, and most people click Connect without a second thought about what the AI can actually do once it is in.
Here is the part people miss. A connector does not just read your stuff. It can act. It can send, delete, move, and change things for you. That is what makes it so useful, and it is also the exact reason you want to decide the rules up front instead of finding out later.
This is part 4 of my AI safety series. Here is the good news. Claude does not make this all or nothing. It gives you granular control over exactly what each connection is allowed to do. You just have to go and set it, and almost nobody does.
One Minute, Done
You only have to set this once per tool. A single minute now saves you a lot of second-guessing later, and it means Claude works the way you want, not the way the default decided for you.
LevelsThe Three Levels
For each connection, Claude lets you pick one of three settings for every action it might take. Here is what each one means and when I reach for it.
Level 1
Always Allow
Claude never asks. It just does that action for you. Pick this for the safe, everyday things you do not want to babysit, like reading a file or searching your inbox. Nothing here can cause damage, so getting out of the way is fine.
Level 2
Ask Every Time
Claude checks with you before it acts, every single time. You see exactly what it wants to do, and you say yes or no. Pick this for anything that changes, sends, or shares. It adds one click, and that one click is worth it.
Level 3
Always Block
Claude can never do that action, full stop. Pick this for the things you never want an AI touching on its own, like deleting files or spending money. If it is off the table for you, put it fully off the table.
My SetupHow I Set Mine
Here are a few of my real rules, so you can see how I think about it. Notice the pattern: I let Claude do the safe half of a job freely, and I keep my hand on the part that cannot be undone.
- Read my Google Drive: yes. I am happy letting Claude read my files whenever it needs to. Reading does not change anything, so I keep it on Always Allow.
- Delete files: never. I set deleting to Always Block. There is no reason an AI needs to erase my work on its own, so it simply cannot.
- Draft emails: yes. I love letting Claude write email drafts for me. A draft just sits there until I look at it, so this one is easy to allow.
- Send emails: only with my approval. Sending is set to Ask Every Time. I read every message before it leaves my inbox, no exceptions.
My Rule Of Thumb
Reads can lean toward Always Allow. They only look, they never change a thing. But anything that deletes, sends, or pays belongs on Ask Every Time or Always Block. Those are the actions you cannot take back, so those are the ones you keep control of.
SetupSet It Up In A Minute
This is quicker than people expect. Here is the whole thing.
- Open your connector settings in Claude.
- Click into the specific connector you want to adjust, like Gmail or Google Drive.
- Set each permission one by one to Always Allow, Ask Every Time, or Always Block.
That is it, about a minute per tool. Do this for every tool the moment you connect it, before you let Claude start working. A new connector should not go live until you have told it what it may and may not touch.
The One Rule
The safest AI setup is the one you control. Set your permissions before you connect Claude to anything that holds real data. It takes a minute, and it puts you back in charge of your own tools.