system_prompts_leaks

Free

Extracted system prompts from ChatGPT (GPT-5.5 Thinking), Claude (Opus 4.7, Opus 4.6, Sonnet 4.6, Claude Code), Gemini (3.1 Pro, 3 Flash, Gemini CLI), Grok (4.3 beta), Perplexity, and more. Updated re

AI ChatbotsFreeFree tier

Inputs: textOutputs: text

Type

Open Source

GitHub (39.4k stars)

About system_prompts_leaks

System Prompts Leaks is an open-source GitHub repository that collects and publishes extracted system prompts from various large language models (LLMs). The project aims to provide transparency into the hidden instructions that guide AI assistants like ChatGPT, Claude, Gemini, Grok, and Perplexity. By making these prompts publicly available, the repository enables researchers, developers, and curious users to understand the underlying rules, constraints, and behaviors programmed into these models. The collection is updated regularly to include new versions and additional models as they become available.

Key Features

Extracted system prompts from multiple LLMs including ChatGPT, Claude, Gemini, Grok, and Perplexity

Open-source repository hosted on GitHub

Regular updates to include new model versions and additional AI assistants

Provides transparency into hidden AI instructions

Free to access and use

Pros & Cons

Pros

Free and open-source, accessible to everyone
Provides valuable transparency into AI system instructions
Regularly updated to include new models and versions
Useful for researchers, developers, and AI enthusiasts
Hosted on GitHub with version control and community contributions

Cons

Accuracy of extracted prompts should be verified as they may not be official or complete
Requires familiarity with GitHub to access and use effectively
May not include all models or the most recent versions
Legal and ethical considerations around using extracted prompts should be reviewed

Best For

Researching the design and constraints of popular AI modelsComparing system prompts across different AI assistantsUnderstanding how AI models are instructed to behaveEducational purposes for AI transparency and ethics studiesDeveloping or fine-tuning AI systems by studying existing prompts

Alternatives to system_prompts_leaks

agents-towards-production

End-to-end, code-first tutorials for building production-grade GenAI agents. From prototype to enterprise deployment.

Agent-Reach

Give your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.

A2A

Agent2Agent (A2A) is an open protocol enabling communication and interoperability between opaque agentic applications.

The AI Toolkit for TypeScript. From the creators of Next.js, the AI SDK is a free open-source library for building AI-powered applications and agents

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

agentscope

Build and run agents you can see, understand and trust.

FAQ

What are system prompts?

System prompts are hidden instructions given to AI models that define their behavior, constraints, and personality. They are typically not visible to end users.

How are these prompts extracted?

The extraction methods are not detailed in the repository, but they likely involve techniques such as prompt injection or reverse engineering. The accuracy of extracted prompts should be verified.

Is this repository legal to use?

The legal status of extracting and sharing system prompts may vary by jurisdiction and the terms of service of the respective AI providers. Users should review applicable laws and terms.

How often is the repository updated?

Based on available information, the repository is updated regularly, but the exact frequency is not specified. Users should check the repository for the latest updates.

Can I contribute to the repository?

As an open-source project on GitHub, contributions are likely welcome. Users should refer to the repository's contributing guidelines for details.