This repository collects and shares system prompts for popular large language models like ChatGPT, Claude, and Gemini. It serves as a community-driven resource for researchers, developers, and enthusiasts interested in understanding and potentially leveraging the underlying instructions that guide chatbot behavior.

How It Works

The project functions as a curated collection of extracted system prompts. These prompts are gathered from publicly available sources and user contributions, aiming to document the specific instructions given to LLMs to shape their responses, persona, and adherence to safety guidelines.

Quick Start & Requirements

• No installation or execution required; it's a data repository.
• Access is via browsing the GitHub repository.

Highlighted Details

• Centralized collection of system prompts for major LLMs.
• Community-driven with open PRs for contributions.
• Includes contact information for the maintainer.

Maintenance & Community

• Maintained by asgeirtj.
• Community interaction encouraged via GitHub Discussions.
• Contact: asgeirtj at gmail.com, Discord: asgeirtj, X: https://x.com/asgeirtj, Reddit: https://www.reddit.com/user/StableSable/

Licensing & Compatibility

• The repository itself does not specify a license. Users should exercise caution regarding the use of extracted prompts, as their underlying ownership and usage rights may vary depending on the original LLM providers.

Limitations & Caveats

The prompts are extracted and may not represent the most current or complete set of instructions used by the LLM providers. The legality and ethical implications of using these prompts may vary and are the responsibility of the user.