anti-slop-writing

Review and rewrite prose to remove generic AI-writing patterns and inflated language.

3|Updated Feb 15, 2025
One-click install
npx skills add https://github.com/mark-torres10/ai_tools --skill anti-slop-writing-mark-torres10
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: anti-slop-writing
Source: https://github.com/mark-torres10/ai_tools/tree/main/skills/anti-slop-writing
Command: npx skills add https://github.com/mark-torres10/ai_tools --skill anti-slop-writing-mark-torres10

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? AI-generated and AI-edited prose often reads as generic, inflated, and formulaic, filled with prestige vocabulary, canned emphasis, and decorative rhythm instead of concrete detail. This Skill detects those patterns and rewrites the text so claims are backed by named mechanisms, evidence, and specific detail. ## Core Features & Use Cases - Slop Detection: Flags AI-writing tells such as rule-of-three overuse, negative parallelisms, decorative em-dash clusters, displaced copulas, hedged symmetry, and banned phrases like "In today's rapidly evolving landscape". - Structured Critique Output: Produces verdicts (keep / revise / ask-author / reject) with concrete rewrites, self-detector checks on the rewrite, and a memorable closing line. - Flow and Conclusion Repair: Fixes list-like paragraph flow with hinge sentences and replaces generic thesis endings with carrier-bound conclusions. - Use Case: Paste a draft blog post, README, launch announcement, or talk intro and receive a line-by-line review that identifies which sentences rely on rhythm instead of evidence, plus rewrites that name the actual mechanism. ## Quick Start Review this draft blog post for AI-writing patterns and rewrite the flagged sentences with concrete detail.

Frequently Asked Questions about anti-slop-writing

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I make AI-generated writing sound less generic?

Run the text through detectors for AI-writing tells such as rule-of-three overuse, negative parallelisms, prestige vocabulary, and decorative em-dash clusters. Then rewrite flagged sentences by replacing abstract importance claims with concrete mechanisms, named actors, and evidence.

What are common signs of AI-generated text?

Common signs include phrases like "In today's rapidly evolving landscape", words like delve, tapestry, and pivotal, copula displacement such as "serves as", hedged symmetry like "Whether you're X or Y", and conclusions that could belong to any essay in the category.

Should all em-dashes and short contrast sentences be removed?

No. The skill classifies contrasts as earned, compressed, or decorative based on whether prior prose evidences both sides. Keep earned antithesis and em-dashes that bracket real definitions or asides; only cut decorative clusters and unsupported contrasts.

When should I ask the author instead of rewriting a sentence?

Use the ask-author verdict when a line is improvable but the fix needs a fact the source does not supply, such as a tool name, count, or specific event. Never invent a confident-sounding specific to fill a rewrite slot.

Can this skill review its own rewrites for slop?

Yes. Every rewrite must pass the same detectors applied to the source, including checks for rule-of-three, X-not-Y cadence, banned phrases, and invented facts. A rewrite that would be flagged as source text must be rewritten again or escalated to ask-author.