When you vibe code, do you actually know what your AI agent ran?

Vibe coding works because we stop reviewing every step. We describe what we want, the agent gets on with it, and we look at the result. That speed is the whole point, and I don't think anyone wants to go back to approving every command.

But it means the agent is running commands, reading files, installing packages and calling tools with our own permissions, mostly unwatched. When something goes wrong, the transcript doesn't help much: it's a conversation, not a record of what actually happened on the machine.

For a long time that felt fine to me. Then I started recording every action my own agents took, and looked at a few days of it.

What surprised me wasn't any single command. It was the combinations. Reading a file with credentials in it is normal. Calling a cloud API is normal. The same session doing one and then the other, without me asking for either, is the kind of thing I want to at least notice. Some of what got flagged I could explain right away. A few I'm still going through, which is exactly the point: without the record, I wouldn't even know there was something to look at.

It also changed how I think about review. Reading the final diff tells you what changed in the code. It tells you nothing about what the agent read, ran or sent along the way.

I'm curious how people here handle this:

1. Do you review what your agent ran during a session, or only the final result? What made you choose that?

2. Has an agent ever done something you only discovered later, like deleting files, touching a .env, installing an unexpected package, or making a network call you didn't ask for?

3. If you had a full record of every action, what would you actually want flagged? Where's the line between useful and noise you'd ignore within a week?

4. For anyone vibe coding in a team or on work projects: is there any shared record of what agents did, or is it trust plus code review?

5. Would you give up a little speed for more visibility, or is any friction a dealbreaker?

I'll share more of what I've seen in the replies, but I'd really like to hear how others approach it first.