This project addresses the pervasive issue of "AI slop"—a distinct, often unoriginal visual and copy style emerging in AI-generated products. It provides a multilingual field guide and an agent skill to identify, explain, and remove these stylistic tics. The project benefits developers and designers aiming to create more original, aesthetically refined, and less machine-generated digital products by offering actionable insights and automated assistance.

How It Works

The project consists of two main components: a static, zero-JS-by-default Astro website serving as a multilingual field guide, and a kill-ai-slop agent skill for coding assistants. The field guide catalogues 32 specific "AI slop" tells, each demonstrated with interactive before-and-after examples. The agent skill scans web projects for code-level indicators of these tells, explains their machine-generated nature, and proposes or applies clean fixes. The project adheres to a strict "paper + ink" design system emphasizing hierarchy, scale, and space, deliberately avoiding common AI tropes like gradients, excessive emoji, and glass effects.

Quick Start & Requirements

• Website: Clone the repository, navigate to the website/ directory, run npm install, then npm run dev for local development or npm run build for static deployment. Requires Node.js and npm.
• Agent Skill: Install universally via npx skills add yetone/kill-ai-slop for compatible agents, or manually by copying the skill/ directory into an agent's skills folder. A dependency-free scanner can be run directly using node skill/scripts/scan.mjs path/to/project.
• Links: Field guide and live demo: https://killaislop.com. Skill installation details: https://github.com/yetone/kill-ai-slop/tree/main/skill.

Highlighted Details

• Catalogues 32 specific "AI slop" tells derived from real-world products.
• Multilingual field guide available in English, Chinese, Japanese, and Korean.
• The scanner is dependency-free and operates non-destructively, never editing project files.
• All demonstration examples are rebuilt in plain HTML for educational clarity.

Maintenance & Community

No specific details regarding maintainers, community channels (e.g., Discord, Slack), sponsorships, or roadmap were found in the provided README text.

Licensing & Compatibility

The license type is not explicitly stated in the provided README text, which requires clarification for adoption decisions, especially concerning commercial use or integration with closed-source projects.

Limitations & Caveats

The README does not detail specific limitations, known bugs, or alpha status. The effectiveness of the agent skill relies on the accuracy and comprehensiveness of its detection patterns and remediation playbooks.