# Application Cleanup Method

> Apply accepted application-subtraction findings with isolated phases and snapshot-bound build/test gates. Load from the host cleanup entry point, code-review Step 7c or lifecycle cleanup.

- Skill: `acaprino/application-cleanup-method` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx get-skill add acaprino/application-cleanup-method`
- Raw SKILL.md: https://api.skillmd.com/api/skills/acaprino/application-cleanup-method/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: acaprino (https://skillmd.com/u/acaprino)
- Updated: 2026-11-11
- Page: https://skillmd.com/skills/acaprino/application-cleanup-method

---


# Application cleanup

Input: accepted cleanup findings and severities, target, explicit scope, output
root, exclusively owned isolated workspace, clean tree, targeted-fix delivery,
commit authorization and required build/test commands. Output: phase ledger with
pre-phase SHA, candidate snapshot, gate evidence, commit or verified recovery.

Use the host run-record mechanism for execution preflight and the canonical
gate/recovery procedure. If absent, declare preflight and recovery unavailable.
Reuse recorded authorization; ask only for missing choices.
Read `references/application-cleanup.md` inside this skill. Its subtraction method
is promoted from Step 7c without adding a new auditor or a semantic refactor engine.
Inputs can come from a review or an identified lifecycle plan. When targeted fixes
were unnecessary, record 7b as not applicable; otherwise its accepted changes must
be delivered and committed before subtraction. A `fix` grant without `commit` leaves
this commit-only action open and explains the missing capability before execution.

