AI Slop Cleaner

1---
2name: ai-slop-cleaner
3description: Clean AI-generated code slop with a regression-safe, deletion-first workflow and optional reviewer-only mode
4level: 3
5---
6 
7# AI Slop Cleaner
8 
9Use this skill to clean AI-generated code slop without drifting scope or changing intended behavior. In OMC, this is the bounded cleanup workflow for code that works but feels bloated, repetitive, weakly tested, or over-abstracted.
10 
11## When to Use
12 
13Use this skill when:
14- the user explicitly says `deslop`, `anti-slop`, or `AI slop`
15- the request is to clean up or refactor code that feels noisy, repetitive, or overly abstract
16- follow-up implementation left duplicate logic, dead code, wrapper layers, boundary leaks, or weak regression coverage
17- the user wants a reviewer-only anti-slop pass via `--review`
18- the goal is simplification and cleanup, not new feature delivery
19 
20## When Not to Use
21 
22Do not use this skill when:
23- the task is mainly a new feature build or product change
24- the user wants a broad redesign instead of an incremental cleanup pass
25- the request is a generic refactor with no simplification or anti-slop intent
26- behavior is too unclear to protect with tests or a concrete verification plan
27 
28## OMC Execution Posture
29 
30- Preserve behavior unless the user explicitly asks for behavior changes.
31- Lock behavior with focused regression tests first whenever practical.
32- Write a cleanup plan before editing code.
33- Prefer deletion over addition.
34- Reuse existing utilities and patterns before introducing new ones.
35- Avoid new dependencies unless the user explicitly requests them.
36- Keep diffs small, reversible, and smell-focused.
37- Stay concise and evidence-dense: inspect, edit, verify, and report.
38- Treat new user instructions as local scope updates without dropping earlier non-conflicting constraints.
39 
40## Scoped File-List Usage
41 
42This skill can be bounded to an explicit file list or changed-file scope when the caller already knows the safe cleanup surface.
43 
44- Good fit: `oh-my-claudecode:ai-slop-cleaner skills/ralph/SKILL.md skills/ai-slop-cleaner/SKILL.md`
45- Good fit: a Ralph session handing off only the files changed in that session
46- Preserve the same regression-safe workflow even when the scope is a short file list
47- Do not silently expand a changed-file scope into broader cleanup work unless the user explicitly asks for it
48 
49## Ralph Integration
50 
51Ralph can invoke this skill as a bounded post-review cleanup pass.
52 
53- In that workflow, the cleaner runs in standard mode (not `--review`)
54- The cleanup scope is the Ralph session's changed files only
55- After the cleanup pass, Ralph re-runs regression verification before completion
56- `--review` remains the reviewer-only follow-up mode, not the default Ralph integration path
57 
58## Review Mode (`--review`)
59 
60`--review` is a reviewer-only pass after cleanup work is drafted. It exists to preserve explicit writer/reviewer separation for anti-slop work.
61 
62- **Writer pass**: make the cleanup changes with behavior locked by tests.
63- **Reviewer pass**: inspect the cleanup plan, changed files, and verification evidence.
64- The same pass must not both write and self-approve high-impact cleanup without a separate review step.
65 
66In review mode:
671. Do **not** start by editing files.
682. Review the cleanup plan, changed files, and regression coverage.
693. Check specifically for:
70   - leftover dead code or unused exports
71   - duplicate logic that should have been consolidated
72   - needless wrappers or abstractions that still blur boundaries
73   - missing tests or weak verification for preserved behavior
74   - cleanup that appears to have changed behavior without intent
754. Produce a reviewer verdict with required follow-ups.
765. Hand needed changes back to a separate writer pass instead of fixing and approving in one step.
77 
78## Workflow
79 
801. **Protect current behavior first**
81   - Identify what must stay the same.
82   - Add or run the narrowest regression tests needed before editing.
83   - If tests cannot come first, record the verification plan explicitly before touching code.
84 
852. **Write a cleanup plan before code**
86   - Bound the pass to the requested files or feature area.
87   - List the concrete smells to remove.
88   - Order the work from safest deletion to riskier consolidation.
89 
903. **Classify the slop before editing**
91   - **Duplication** — repeated logic, copy-paste branches, redundant helpers
92   - **Dead code** — unused code, unreachable branches, stale flags, debug leftovers
93   - **Needless abstraction** — pass-through wrappers, speculative indirection, single-use helper layers
94   - **Boundary violations** — hidden coupling, misplaced responsibilities, wrong-layer imports or side effects
95   - **Missing tests** — behavior not locked, weak regression coverage, edge-case gaps
96 
974. **Run one smell-focused pass at a time**
98   - **Pass 1: Dead code deletion**
99   - **Pass 2: Duplicate removal**
100   - **Pass 3: Naming and error-handling cleanup**
101   - **Pass 4: Test reinforcement**
102   - Re-run targeted verification after each pass.
103   - Do not bundle unrelated refactors into the same edit set.
104 
1055. **Run the quality gates**
106   - Keep regression tests green.
107   - Run the relevant lint, typecheck, and unit/integration tests for the touched area.
108   - Run existing static or security checks when available.
109   - If a gate fails, fix the issue or back out the risky cleanup instead of forcing it through.
110 
1116. **Close with an evidence-dense report**
112   Always report:
113   - **Changed files**
114   - **Simplifications**
115   - **Behavior lock / verification run**
116   - **Remaining risks**
117 
118## Usage
119 
120- `/oh-my-claudecode:ai-slop-cleaner <target>`
121- `/oh-my-claudecode:ai-slop-cleaner <target> --review`
122- `/oh-my-claudecode:ai-slop-cleaner <file-a> <file-b> <file-c>`
123- From Ralph: run the cleaner on the Ralph session's changed files only, then return to Ralph for post-cleanup regression verification
124 
125## Good Fits
126 
127**Good:** `deslop this module: too many wrappers, duplicate helpers, and dead code`
128 
129**Good:** `cleanup the AI slop in src/auth and tighten boundaries without changing behavior`
130 
131**Bad:** `refactor auth to support SSO`
132 
133**Bad:** `clean up formatting`
134