we use a service called Briefhq.

First: It’s a MCP/cli you can hook up to Claude code and slack, it integrates with GitHub.

the harness lets you record decision as contextual info you can pull up whenever your start a planning session.

It also makes sure your decisions don’t conflict with each other.

I find myself talking through a decision I made months ago with it, update it with any new decisions and it just figures out how to merge everything.

No extra workflow outside of this.

_Are you letting AI interact directly with ops tools?_

Yes but very cautiously and always with human in the loop. AI is great at cutting away ops noise (false alerts), OK at root cause analysis / debugging and terrible at actually prioritizing and fixing the thing.

It's not a solved a problem and companies like Rootly's actively speaking in this area. I'd read their blog and see what's the latest in the field.

why do this instead of using a merchant of record?

Well they did and Facebook refused. Hence the "do not track" feature.