no-ai-slop

Rewrites prose to remove AI-generated writing patterns and unsupported claims.

Updated May 18, 2026
One-click install
npx skills add https://github.com/blalor/pi-dot-dev --skill no-ai-slop-blalor
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: no-ai-slop
Source: https://github.com/blalor/pi-dot-dev/tree/main/agent/skills/no-ai-slop
Command: npx skills add https://github.com/blalor/pi-dot-dev --skill no-ai-slop-blalor

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? AI-generated drafts contain recognizable patterns such as em dashes, intensifiers, hollow claims, stock transitions, and fabricated attributions that make prose sound mechanical and untrustworthy. This Skill provides a concrete rule set and self-check process for producing clear, direct, evidence-based writing. ## Core Features & Use Cases - 24 writing rules: Covers banned phrases, inflated verbs, academic filler, heading anti-patterns, fabricated attributions, and synthetic enthusiasm, each with bad and better examples. - Expanded detection reference: The bundled references/ai-writing-detection.md file adds word lists, hedging thresholds, structural uniformity checks, and model-family-specific tells for deeper lint passes. - Self-check workflow: An 11-step checklist verifies numbers against sources, removes citation artifacts, and confirms headings describe their sections. - Use Case: Before publishing a technical blog post drafted with AI assistance, run the rules to replace claims like "significantly faster" with measured benchmark values and rewrite dramatic headings into descriptive ones. ## Quick Start Review this draft against the no-ai-slop rules and rewrite any sentences that violate them.

Frequently Asked Questions about no-ai-slop

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

FAQPage Schema
How do I make AI-generated text sound more human?▼

Remove em dashes, intensifiers, filler openings, and stock transitions, then replace hollow claims with concrete facts. Vary sentence and paragraph lengths, and read the text aloud to catch phrasing that sounds unnatural in conversation.

What words and phrases signal AI-generated writing?▼

Common tells include delve, leverage, facilitate, furthermore, moreover, robust, comprehensive, and phrases like "in today's world" or "it's important to note." The references file lists overused verbs, adjectives, transitions, and inflated symbolism phrases with direct replacements.

How do I check text for AI writing patterns before publishing?▼

Follow the 11-step self-check: search for em dashes, verify every number and quotation against a source, remove filler and inflated verbs, and confirm headings describe their sections. For publication, run the expanded lint pass in references/ai-writing-detection.md.

Does this writing style apply to code, quotes, and technical terms?▼

No. The rules explicitly exclude required terminology, exact quotations, code, configuration, product names, and factual work logs. Quoted text must remain verbatim, with any changes marked in square brackets.

When should I not apply these writing rules?▼

Do not apply them when the artifact is a research log, validation report, or limitations section, where documenting unsuccessful searches and missing evidence is the point. Hedging is also appropriate for genuinely uncertain or disputed facts.