audit-decisions
Audit Decisions
Run this only when a person asked for it by name, or a brief or workflow you are executing names it (that is the only unsupervised use). If you loaded it on your own, stop and ask first: it writes to the PR.
Given a good decision, an agent implements it faithfully; wherever the task is underspecified, it makes the decision itself — silently, and the diff won't flag it. Reviewing thousands of changed lines doesn't scale, and it inspects the execution, which was probably fine. The audit that scales is of the decisions: surface every one the implementer made on its own, judge that list, and record the verdicts.
For us the implementer is the lane, and the "original spec or prompt" is the lane brief — a decision the brief did not cover is the one under audit.
This is about architecture more than bugs. An implementation can work perfectly and still rest on decisions the user never made — a data shape, a storage location, a dependency, an API contract, a tradeoff of memory for speed — and every one of them is load-bearing for future work. The user needs to know them not because they're wrong, but because they now own them.