Awesome LLM System Prompts 🤖

A curated collection of system prompts from various Large Language Models (LLMs), including official releases and community discoveries.

📋 Table of Contents

• About
• Repository Structure
• Models Covered
• Contributing
• Disclaimer
• License

About

This repository aims to document and preserve system prompts from various LLMs to:

• Foster transparency in AI systems
• Enable research and analysis
• Help developers understand prompt engineering techniques
• Create a historical record of prompt evolution

Repository Structure

├── README.md
├── CONTRIBUTING.md
├── LICENSE
├── example-prompt-file.md
├── models/
│   ├── anthropic/
│   │   ├── claude-3-opus/
│   │   ├── claude-3-sonnet/
│   │   ├── claude-3-haiku/
│   │   └── claude-4-sonnet/
│   ├── openai/
│   │   ├── gpt-3.5-turbo/
│   │   ├── gpt-4/
│   │   ├── gpt-4-turbo/
│   │   └── o1-preview/
│   ├── google/
│   │   ├── gemini-pro/
│   │   ├── gemini-ultra/
│   │   └── bard/
│   ├── meta/
│   │   ├── llama-2/
│   │   └── llama-3/
│   ├── microsoft/
│   │   └── copilot/
│   └── other/
├── analysis/
│   ├── prompt-patterns.md
│   ├── evolution-timeline.md
│   └── comparative-analysis.md
└── tools/
    ├── prompt-extractor.py
    └── analysis-scripts/

Models Covered

🔵 Anthropic

• Claude 4 Sonnet - Latest version (August 2025)
• Claude 3 Opus - Most capable model
• Claude 3 Sonnet - Balanced performance
• Claude 3 Haiku - Fast and efficient

🟢 OpenAI

• GPT-4 Turbo - Enhanced version
• GPT-4 - Original flagship model
• GPT-3.5 Turbo - Widely used variant
• O1-Preview - Reasoning-focused model

🔴 Google

• Gemini Ultra - Most capable
• Gemini Pro - General purpose
• Bard - Conversational AI

🔵 Meta

• Llama 3 - Latest open model
• Llama 2 - Previous generation

🟡 Microsoft

• Copilot - Integrated assistant

Others

• Mistral AI models
• Cohere models
• Perplexity AI
• Community fine-tunes

File Naming Convention

Each prompt file follows this naming pattern:

[model-name]_[version]_[date]_[source].md

Examples:

• claude-4-sonnet_20250819_official.md
• gpt-4-turbo_20240315_leaked.md
• gemini-pro_20240201_reverse-engineered.md

Metadata Format

Each prompt file includes this metadata header:

---
model: "Claude 4 Sonnet"
version: "20250819"
source: "official"
date_discovered: "2025-08-19"
verified: true
language: "en"
context_length: "200k"
tags: ["assistant", "helpful", "harmless", "honest"]
---

Contributing

We welcome contributions! Please see CONTRIBUTING.md for guidelines.

How to Contribute

1. Fork the repository
2. Create a new branch for your contribution
3. Add the system prompt with proper metadata
4. Verify the prompt's authenticity when possible
5. Submit a pull request

Contribution Types

• ✅ Official prompts from model providers
• ✅ Reverse-engineered prompts with evidence
• ✅ Leaked prompts with verification
• ✅ Analysis and comparative studies
• ❌ Speculation without evidence
• ❌ Proprietary content without permission

Verification Levels

• 🟢 Verified Official - Confirmed by model provider
• 🟡 Community Verified - Multiple independent confirmations
• 🟠 Probable - Strong evidence but not fully confirmed
• 🔴 Unverified - Requires additional verification

Analysis Tools

The repository includes tools for:

• Prompt extraction and parsing
• Similarity analysis between models
• Evolution tracking over time
• Pattern identification
• Statistical analysis

Research Applications

This collection enables research in:

• Prompt Engineering - Understanding effective techniques
• AI Safety - Analyzing safety instructions and constraints
• Model Behavior - Correlating prompts with outputs
• Evolution Studies - Tracking changes over time
• Comparative Analysis - Understanding different approaches

Ethical Considerations

• All prompts are collected for educational and research purposes
• We respect intellectual property rights
• Sources are properly attributed
• Sensitive information is redacted when necessary

Disclaimer

This repository is for educational and research purposes only. System prompts may be:

• Incomplete or outdated
• Subject to change by model providers
• Protected by intellectual property rights
• Not representative of current model behavior

Always verify information independently and respect terms of service.

License

This project is licensed under MIT License - see LICENSE file for details.

Contact

• Issues: Use GitHub issues for questions or problems
• Discussions: Use GitHub discussions for general conversation
• Email: [your-email@example.com]

---

⭐ Star this repository if you find it useful for your research or development work!

Recent Updates

• 2025-08-19: Added Claude 4 Sonnet system prompt
• 2025-08-15: Improved verification methodology
• 2025-08-10: Added comparative analysis tools
• 2025-08-05: Repository restructure for better organization

Statistics

• Total Models: 25+
• Total Prompts: 150+
• Contributors: 50+
• Last Updated: August 2025