This repository provides a curated collection of instruction-based prompts for large language models like ChatGPT, GPT-3.5, and GPT-4. It aims to help users effectively guide these models for various text generation, classification, and analysis tasks, serving as a practical resource for developers, researchers, and content creators.

How It Works

The core principle is leveraging specific "instruction verbs" as the initial command to steer the LLM's output. By combining these verbs with clear directives and contextual information, users can elicit precise and desired responses. The prompts are categorized by task type (classification, generation, transformation, comparison, etc.) and demonstrate effective prompt structuring, including the use of explicit instructions and optional conversational phrasing.

Quick Start & Requirements

• Usage: Copy and paste prompts directly into LLM interfaces (e.g., ChatGPT, API).
• Requirements: Access to a compatible LLM (e.g., OpenAI's GPT models). No specific software installation is needed beyond LLM access.

Highlighted Details

• Comprehensive list of instruction verbs for diverse NLP tasks.
• Examples demonstrating prompt structure and effective phrasing.
• Categorization of prompts by common use cases like classification, generation, and transformation.
• Guidance on advanced prompting techniques (zero-shot, few-shot, chain-of-thought).

Maintenance & Community

• Project maintained by Kevin Amiri.
• No specific community links (Discord, Slack) or roadmap are provided in the README.

Licensing & Compatibility

• The README does not explicitly state a license.

Limitations & Caveats

The effectiveness of prompts is highly dependent on the specific LLM used and its underlying architecture. The collection is a resource for prompt engineering rather than a runnable software project.