whereismyai
Leaked System Prompt AI
Leaked System Prompt AI is an open source research repository documenting system prompts and instructions from major AI models.
This project collects and documents system prompts, behavioral instructions, and tool definitions from major AI models and agent platforms. It aims to make AI systems more transparent, understandable, and observable by studying the instructions behind model outputs. The repository covers platforms such as OpenAI, Google, Anthropic, xAI, Perplexity, Cursor, Windsurf, Devin, Manus, and Replit. Entries contain model names, versions, dates, system prompts, guidelines, tools, configurations, and sources. The project is intended for research, education, analysis, and documentation.
Leaked System Prompt AI at a glance
Key features
- Collects system prompts from major AI models
- Documents behavioral instructions and guidelines
- Catalogs tool definitions and configurations
- Preserves historical records of AI configurations
- Provides context for AI research and analysis
Use it for
- Studying how AI systems are instructed and configured
- Analyzing safety and policy frameworks of AI models
- Researching behavior and personas of AI agents
- Documenting system prompts from major platforms
Alternatives to Leaked System Prompt AI
- Promptfoo — It simulates real users to uncover agent and RAG vulnerabilities.
- deepsense.ai — A framework to evaluate agentic systems before production usage exists.
- Agent Cockpit — It checks evidence chains deterministically so broken claims never slip by unnoticed.
- Beeuni — Agents retain context and improve skills over time through a controlled review process.
- Oracle — Provides a single entry point for multiple pro AI agents and session management.
- Kungfu — Maintains work state and context when switching between different AI agents.
- Jonex — Provides traceable, agent-ready knowledge from diverse enterprise data sources.
- Entelligence AI — Inspect agent sessions locally without digging through logs or sharing data.
- Reflexio — It autonomously extracts lessons from interactions to permanently correct agent behavior.
- PromptHub — It enables teams to manage, version, and evaluate prompts in a centralized UI.
- Grim Prompter — It guides AI coding agents to produce evidence-traced threat models for repositories.
- GitAgent — Manage AI agents with familiar git workflows like branching, diffing, and reverting.
All Leaked System Prompt AI alternatives