<2-3 sentences summarising what the project addresses and solves, the target audience, and the benefit.> This repository provides a framework to combat "AI slop" by enforcing rules and a specific voice profile to make AI-generated text sound more human and less machine-like. It targets users of AI writing tools who want to improve the quality, trustworthiness, and readability of their AI-assisted content, ensuring it passes human scrutiny. The benefit is more authentic and credible written output.

How It Works

The project provides two core components for AI models: a strict set of 24 "anti-slop" rules to eliminate machine-generated text patterns (e.g., filler phrases, dramatic headings, specific sentence structures) and a data-driven voice profile derived from over 500,000 words of Louis Rossmann's writing. This profile guides the AI on sentence variation, paragraph structure, and a precise, evidence-based argumentative style. The AI reads these directives before generating or editing prose and then self-checks its output against a banned-words reference.

Quick Start & Requirements

• Primary install / run command: Integrate the provided files into an AI model's workflow (e.g., by dropping the folder next to a project or pointing Claude Code at it).
• Non-default prerequisites and dependencies: Requires an AI model capable of processing custom instructions or skills (e.g., Claude Code). No other specific software dependencies are listed.
• Estimated setup time or resource footprint: Minimal setup, primarily involving integration into the AI's operational context.
• Links: The README serves as the primary documentation; no separate quick-start or demo links are provided.

Highlighted Details

• 24 Anti-Slop Rules: Targets common AI writing tics like emdashes, intensifiers, filler phrases, and dramatic headings.
• Rossmann Voice Profile: Captures stylistic elements from 513,683 words, including number density, sentence-length variance, contractions, and a "contempt through precision" approach.
• Self-Correction: AI output is validated against a banned-words reference before delivery.
• General Applicability: Rules and voice profile are designed for essays, scripts, documentation, emails, and any text-based content.

Maintenance & Community

• Primarily associated with the realrossmanngroup.
• No specific community channels (Discord, Slack, etc.) or roadmap links are mentioned in the README.

Licensing & Compatibility

• License type and notable restrictions: The repository's license is not specified in the README, hindering compatibility assessment for commercial or closed-source use.
• Compatibility notes: Designed for use with AI models that support custom rule sets and voice profiles.

Limitations & Caveats

• The voice profile's example corpus is repair-themed, though traits are claimed general-purpose.
• Crucially, the repository's license is not specified in the README, hindering compatibility assessment for commercial or closed-source use.
• Lacks explicit community channels or roadmap details.