The Agent Checklist That Actually Lets Me Sleep

After the week of rogue-agent stories, I stopped asking whether a model was smart enough to deploy and started asking whether I could prove what it did after. The second question is the one that keeps me safe.

My checklist is short and non-negotiable. List every action the agent can take that can't be undone. Gate each one behind a human or a hard rule. Keep an immutable log of what ran. None of it is exotic, and all of it is boring, which is why people skip it.

The part that changed my behavior is the log. An agent I can't explain after the fact is a liability wearing a productivity badge. If I can't audit it, it doesn't touch anything real, full stop.

I saw teams treat safety as a model feature, something the lab ships. It isn't. The safe agent is an architecture: scoped permissions, an allowlist, a human gate on the irreversible. The model is just the part that talks.

Deploy the capability, install the cage. The cage is the feature, and skipping it isn't speed. It's a bet you'll regret the first time the agent does something you didn't ask for.