A Constraint You Can Quietly Lift Isn't One
I had a harness measuring whether a set of skills helped a cheap model. It came back negative: no lift worth the tokens. I believed it.
I had a harness measuring whether a set of skills helped a cheap model. It came back negative: no lift worth the tokens. I believed it.
The reviewer is the visible product: it reads a diff, applies a rule corpus tested against real PRs and code, and ships the findings worth a person's attention.
I built an eval to tell me whether my agent's skills were earning their place. For a while it said they weren't — clean, green, confident.
I asked the agent to review a diff. Two skills woke up and started fighting over whose job it was.
You wrote a skill. It's good. Now you want it to lean on another skill — call it, require it, ship the two together so installing yours pulls in the one it…