PROMPTS

PROMPTS is a comprehensive collection of over 100 AI prompting strategies, curated and documented in German. This project serves as a high-quality resource for German-speaking users to explore, understand, and apply advanced prompting techniques for Large Language Models (LLMs).

The project is built as a static site generator using Bun, ensuring fast builds and a lightweight footprint. It features a semantic ontology (RDF) for machine-readability and a clean, responsive frontend for easy browsing.

Features

• ** German Language Content**: All prompts and strategies are written in German, catering specifically to the DACH region and German-speaking AI enthusiasts.
• ** Extensive Library**: Contains 100+ distinct prompting strategies, range from basic techniques to advanced methods like Chain-of-Thought, ReAct, and Tree of Thoughts.
• ** Fast Static Generation**: Powered by Bun, the build process parses Markdown content and generates a static HTML website instantly.
• ** Semantic Data**: Automatically generates RDF (Resource Description Framework) data, exporting the collection as Turtle .ttl and JSON-LD, making the strategies machine-readable and semantically linked.
• ** Search & Discovery**:
  • Search: Real-time search by title and tags.
  • Filtering: Filter strategies by difficulty (Beginner, Intermediate, Advanced).
  • Related Strategies: Automatically suggests related prompts based on tag overlap.
• ** Modern Frontend**:
  • Minimalist, distraction-free design.
  • Dark/Light mode support.
  • Tag-based navigation.

Project Structure

• content/: The heart of the project. Contains over 100 Markdown files (.md), each representing a specific prompting strategy.
  • Each file includes YAML frontmatter for metadata (Title, Difficulty, Tags) and the strategy content.
• build.ts: The main build script. It reads the content/ directory, parses Markdown and Frontmatter, and generates the dist/ directory.
• rdf.ts: Handles the generation of semantic data (RDF/Turtle) for the collection.
• serve.ts: A simple development server to preview the static site locally.
• src/: Contains frontend assets like template.html, style.css, and script.js.
• dist/: The output directory where the static site is generated.

Getting Started

Prerequisites

• Bun (v1.0 or higher) must be installed on your machine.

Installation

Clone the repository and install dependencies:

git clone <repository-url>
cd prompts
bun install

Usage

1. Build the Project: Generate the static website from the markdown content.

   bun run build.ts

   This will populate the dist/ folder with index.html, semantic data files, and assets.

2. Run Locally: Start a local server to preview the site.

   bun run serve.ts

   Open your browser at http://localhost:3000.

License

MIT License (or clear state if unknown)