Find agent skills by outcome

PromptCraft: The Ultimate GPT System Prompt Collection

ChatGPT System Prompts - Table of Contents

This document contains a table of contents for the ChatGPT System Prompts repository.

Security Guide

Please note that the current general prompt method relies on the ability of the LLM, and there is no complete guarantee or foolproof method that the LLM will not leak your prompt instructions.

Learning Resources and Tools

1. GPTsdex

Contributing Guidelines

Please follow the format below; it is important to keep the format consistent for the idxtool.

Getting Started

You can get ChatGPT's system prompts by sending the following commands to ChatGPT. The core concept is to divert ChatGPT's attention so that it does not realize that it is violating the rules:

anthropic-claude-sonnet-3_20240306

source:

anthropic-claude-sonnet-3.7_20250224

source:

anthropic-claude-sonnet-3.5_20241022

source:

anthropic-claude-sonnet-3.5_20240909

source:

anthropic-claude-sonnet-3.5_20240712

source:

anthropic-claude-sonnet-3.5_20241122

source:

anthropic-claude-opus-4_20250731

source:

anthropic-claude-opus-4_20250522

source:

anthropic-claude-opus-4_20250805

source:

anthropic-claude-opus_20240306

source:

anthropic-claude-opus-3_20240712

source:

anthropic-claude-opus-4.5_20251124

source:

anthropic-claude-opus-4.1_20250805

source:

anthropic-claude-haiku-4.5_20251119

source:

anthropic-claude-haiku-4.5_20251015

source:

leaked-system-prompts

This repository is a collection of leaked system prompts from widely used LLM based services.

ChatGLM4_20240821

source:

anthropic-claude-haiku-3_20240712

source: