// Remove telltale AI writing patterns from documentation and text. Direct, concise prose without AI artifacts.
| name | de-slopify |
| description | Remove telltale AI writing patterns from documentation and text. Direct, concise prose without AI artifacts. |
Read through the complete text carefully and look for any telltale signs of "AI slop" style writing.
Em dashes: One of the biggest tells. Replace with a semicolon, a comma, or recast the sentence so it sounds natural while avoiding em dashes entirely.
AI-isms: Sentences of the form "It's not [just] XYZ, it's ABC" or "Here's why" or "Here's why it matters:" or "Let's dive in." Anything that sounds like the kind of thing an LLM would write disproportionately more commonly than a human writer and which sounds inauthentic or cringe.
Filler openers: "Certainly!", "Great question!", "Absolutely!", "I'd be happy to help!", "Sure thing!" Remove these entirely.
Hedging: "I think maybe", "It might be possible that", "It's worth noting that." Be direct instead.
Excessive bullets: Where flowing prose would work better. Not every list needs bullet points.
Gratuitous emoji: Unless the original text intentionally uses emoji for a specific purpose.
Meta-commentary: "As an AI", "In my experience as a language model", "I was trained to." Remove all self-referential AI commentary.
Redundant transitions: "Furthermore", "Moreover", "Additionally" used as paragraph openers when the connection is already clear.
You cannot do this using regex or a script. You MUST manually read each line of the text and revise it manually in a systematic, methodical, diligent way.
Read the output aloud (mentally) to catch unnatural phrasing.
Apply to the specified file, selection, or most recent output. If no target is specified, ask what text to de-slopify.
Based on Jeffrey Emanuel's De-Slopifier (@doodlestein)
Multi-model code review for gastown. Reviews GitHub PRs (by URL or number) or local branches (by branch name). Spawns parallel Claude, Codex, and (optionally) Gemini reviewers with specialized prompts optimized for regression prevention, then synthesizes findings into a single maintainer-grade decision report. Use --skip-gemini for dual-model mode when Gemini quota is exhausted. Invoke with /review-pr <pr-url|number|branch>.
Multi-pass code review: correctness, security, performance, and maintainability. Each finding gets severity, location, and suggestion.
Structured brainstorming: generate 30 ideas, critically evaluate each, distill to the best 5 with actionable implementation plans.
Multi-phase planning process: understand, decompose, design, validate, then implement. No code until the plan is approved.
Evaluate and polish README and documentation. Ensure accuracy, structure, runnable examples, then de-slopify.
Design and implement an agent-optimized CLI interface for any project. JSON output, structured errors, exit codes, token-efficient.
