# Cleanup > Bidirectional AI slop scanner — Claude + GPT independently analyze, then debate disagreements. Use when this capability is needed. - Skill: `tomevault-io/cleanup-5` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/cleanup-5` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/cleanup-5/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Coding & Dev Tools - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/cleanup-5 --- # /cleanup — Bidirectional AI Slop Scanner ## Usage `/cleanup [scope]` where scope is: deps, unused-exports, hardcoded, duplicates, deadcode, or all ## Description Claude analyzes independently, then codemoot cleanup runs deterministic regex + GPT scans. 3-way merge with majority-vote confidence. ## Instructions ### Phase 1: Claude Independent Analysis Scan the codebase yourself using Grep/Glob/Read. For each scope: - **deps**: Check package.json deps against actual imports - **unused-exports**: Find exported symbols not imported elsewhere - **hardcoded**: Magic numbers, URLs, credentials - **duplicates**: Similar function logic across files - **deadcode**: Declared but never referenced Save findings as JSON to a temp file. ### Phase 2: Run codemoot cleanup ```bash codemoot cleanup --scope SCOPE --host-findings /path/to/claude-findings.json ``` ### Phase 3: Present merged results Show summary: total, high confidence, disputed, adjudicated, by source. ### Phase 4: Rebuttal Round For Claude/GPT disagreements, optionally debate via `codemoot debate turn`. --- > Converted and distributed by [TomeVault](https://tomevault.io/claim/katarmal-ram) — claim your Tome and manage your conversions.