| name | ai-slop-detector |
| description | Detect and remediate AI-generated slop in copywriting/text, frontend UI/design, React components, and backend/architecture code. Use this skill whenever reviewing code before shipping, auditing AI-generated content, doing code reviews, evaluating copywriting, or when asked to improve text/UI/code that looks or feels AI-generated. Trigger on phrases like: does this look AI-generated, clean up this copy, this looks like AI slop, review this code for quality, make this less AI, improve this UI, audit for vibe coding anti-patterns, this copy sounds robotic, or any request to review or improve AI-assisted output across text, UI, or code.
|
AI Slop Detector & Remediator
A comprehensive skill for detecting AI-generated "slop" across four domains — copywriting/text, UI/design, React/frontend code, and backend/architecture — and providing actionable, prescriptive fixes with before/after examples.
Quick Reference: Domain Index
Always load the relevant reference file(s) before diagnosing or fixing.
Workflow
Step 1 — Identify Domain(s)
Determine what's being reviewed: text, UI/CSS/HTML, React, backend, or a mix.
Step 2 — Load Reference File
Read the appropriate reference .md file(s). For cross-domain reviews, load all relevant ones.
Step 3 — Scan & Flag
Work through the content systematically. For each slop signal:
- Name the specific pattern (use the pattern names from the reference files)
- Quote or point to the exact location
- Severity: 🔴 Critical (ships broken/insecure/embarrassing) / 🟡 Medium (hurts quality/trust) / 🟢 Minor (polish/style)
Step 4 — Remediate
For each flagged issue provide:
- Why it's slop — one-sentence explanation
- Before — the problematic excerpt (quoted or referenced)
- After — a concrete rewrite or fix
- Principle — the rule being violated
Step 5 — Summary Report
End every audit with:
- Slop density: Low / Medium / High / Severe
- Top 3 highest-impact fixes (prioritized by severity then effort)
- Full checklist of all issues found, ordered by severity
Universal Slop Signals (All Domains)
These appear across text, code comments, and UI copy alike:
- Over-hedging: "It's worth noting...", "It's important to understand...", "Please note..."
- Fake depth: Claims complexity/nuance/comprehensiveness without demonstrating it
- Structural uniformity: Every section identical in length/format — AI loves symmetry, humans don't
- Missing failure modes: No error states, edge cases, loading states, or empty states
- Passive voice authority: "It is considered best practice" instead of owning a position
- Unnecessary preamble: Restating the question before answering it
- The not-just-X-but-Y construction: "Not just a tool, but a partner" — statistically overrepresented in AI text
- Excessive caveat stacking: Multiple hedges in the same sentence dilute every claim
Scripts
For programmatic/CI use, see:
Run any script with node <script> --help for usage.
