slop-scanlisted

# this skill Repo-wide slop scanner. Audits every tracked file at HEAD for AI slop, verbose code/comments, fragile hard-coded references, and low-value contributions. Optionally scans recent commit history (advisory only). With `--apply`, each finding becomes its own forward-going commit, after the project's discovered quality gates pass. This command **never rewrites history**. It only adds new commits at HEAD. It is safe on pushed branches and cannot flatten merge topology. For branch-scoped slop cleanup that uses fixup commits and `git rebase -i --autosquash`, see the sibling the `pr-deslop` skill in the pr plugin. This is a slash command, not a model-invocable skill: it modifies files and creates commits, so it must be user-explicit, not router-inferred. ## Core thesis Slop is a workflow label, not proof the text is wrong. The skill's job is to reduce review-hostile noise in the codebase, not to scrub the author's voice. Three disciplines: 1. **Audit-first; commits-on-explicit-opt-in.** Default produces a patch-series review surface. `--apply` is the explicit opt-in for landing commits. 2. **Three severity tiers govern auto-apply.** Tier A (deterministic, near-zero FP) auto-applies. Tier B (high-confidence regex with edge cases) is user-confirmed per finding. Tier C (subjective tone) is advisory only and **calibrated against the project's accepted voice on `origin/<trunk>`**, not HEAD. 3. **One finding, one commit; quality gates first.** Every com