repo-hygienelisted

# Repo hygiene ## Scope and triggers Also fires on "audit repo hygiene", "find the stale files". An audit-loop instance (docs/patterns/audit-loop.md). Enumerates files inside a git repo that have decayed (unreferenced docs, stale staging drops, records parked in a control surface, logs past their documented budget, large cold gitignored dirs), verdicts each with evidence inline, co-locates mis-shelved records on a branch, and gates through a PR. The machine surface outside a git repo (abandoned home-folder tool dirs, package caches) belongs to ops-toolkit tools/disk-reclaim, which owns read-first machine cleanup. ## Overview Audit one git repo for files that have decayed and ship the result as a PR. This is the repo-hygiene instance of `docs/patterns/audit-loop.md`. It is the first instance whose item set is the repo's own FILES rather than their claims: `doc-drift` asks whether a doc still tells the truth, this asks whether the file should still be sitting where it is. Two constraints shape everything below, both learned running this pass by hand: - **Every finding carries its evidence inline.** Verification, not detection, was about 90 percent of the manual pass's cost. A findings list without evidence only moves that cost. - **The loop SURFACES, it never deletes.** Route versus trash needed a human call four times in one pass, and twice the human chose against the recommendation. The only fix this instance applies is a MOVE, and even a move gates through the PR