# How do you decide if a PowerShell script from GitHub is safe to run against your tenant? — retained record

> Retained reddit record from reddit:AZURE.

Canonical: https://highsignal.app/data/records/ffc4d0646aca71df

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:AZURE)  published 2026-10-28 02:26 ingested 2026-10-29 02:41

[original source ↗](https://www.reddit.com/r/AZURE/comments/1xhc8cc/how_do_you_decide_if_a_powershell_script_from/)

## retained text



```
Genuine question for people doing Entra / M365 security work.

When you find a script online, say something to pull inactive users or check app permissions, what do you actually do before running it? Read every line? Try it in a test tenant first? Only trust known authors? Or just run it and hope for the best?

I keep running into this. Half these scripts start with Connect-MgGraph asking for a pile of scopes and I can’t always tell if they really need all of them, or what they’ll do with them once they have them.

Has a script ever done something you didn’t expect? And does your team have any actual rule about this, or is it on whoever’s running it?

Full disclosure, I’m working on a side project around this, but not linking anything. Mostly trying to figure out if this is a real pain or just me being paranoid.
```



document fetched 2026-10-29 00:40 · canonical https://reddit.com/r/AZURE/comments/1xhc8cc/how_do_you_decide_if_a_powershell_script_from
