Open Source Project of the Day (Part 51): system-prompts-and-models-of-ai-tools - Resource Collection of AI Tool System Prompts and Models

A deep dive into system-prompts-and-models-of-ai-tools, a resource collection featuring system prompts, internal tools, and AI models from 30+ mainstream AI tools including Cursor, Claude Code, Windsurf, Devin AI, v0, and more, with over 30,000+ lines of system prompt insights

·11 min read·Resource Recommendations
Open Sourcesystem-promptsAI ToolsSystem PromptsCursorClaude CodeWindsurfDevin AIv0Resource Collection

Introduction

"Over 30,000+ lines of insights into their structure and functionality."

This is Part 51 of the "Open Source Project of the Day" series. Today we explore system-prompts-and-models-of-ai-tools (GitHub).

Want to understand how mainstream AI tools like Cursor, Claude Code, Windsurf, and Devin AI work? What do their system prompts look like? What internal tools and AI models do they use? system-prompts-and-models-of-ai-tools is a resource collection project that gathers system prompts, internal tools, and AI models from 30+ mainstream AI tools, including Augment Code, Claude Code, Cluely, CodeBuddy, Comet, Cursor, Devin AI, Junie, Kiro, Leap.new, Lovable, Manus, NotionAI, Orchids.app, Perplexity, Poke, Qoder, Replit, Same.dev, Trae, Traycer AI, VSCode Agent, Warp.dev, Windsurf, Xcode, Z.ai Code, Dia, v0, and more, with over 30,000+ lines of system prompt insights, helping developers understand how these tools work, learn best practices, and research AI Agent architecture design.

Why is it worth exploring?

  • 📚 Resource Collection: Gathers system prompts and model information from 30+ mainstream AI tools
  • 🔍 Deep Insights: Over 30,000+ lines of system prompt analysis revealing how tools work
  • 🎯 Learning Value: Helps developers understand AI tool design principles and best practices
  • 🔬 Research Value: Provides references and inspiration for AI Agent architecture design
  • 🌟 High Popularity: 131k+ Stars, 33.3k+ Forks, highly recognized by the community
  • 🔄 Continuously Updated: Project continuously updated, tracking latest AI tools

What You'll Learn

  • system-prompts-and-models-of-ai-tools positioning: Value of resource collection projects
  • Tool coverage: System prompts from 30+ mainstream AI tools
  • System prompt analysis: How to understand tool principles from system prompts
  • Learning value: How to leverage these resources for learning and research
  • Project structure: Organization and update mechanisms of resource collections
  • Community contributions: How to participate in project contributions and maintenance

Prerequisites

  • Understanding of AI tools basics (Cursor, Claude Code, Windsurf, etc.)
  • Understanding of System Prompts basics
  • Understanding of AI Agent basic architecture (tools, models, prompts)
  • Understanding of resource collection projects characteristics (Awesome List, resource libraries, etc.)

Project Background

Project Overview

system-prompts-and-models-of-ai-tools is a resource collection project aimed at collecting and organizing system prompts, internal tools, and AI model information from mainstream AI tools. This information is typically not publicly available, but through reverse engineering, official documentation, and open source code, project maintainers have collected and organized these valuable resources.

Core Value:

  1. Learning Value: Helps developers understand how AI tools work and their design principles
  2. Research Value: Provides references and inspiration for AI Agent architecture design
  3. Educational Value: Helps beginners understand internal mechanisms of AI tools
  4. Transparency: Promotes transparency and understandability of AI tools

Project Characteristics:

  • Comprehensive: Covers 30+ mainstream AI tools
  • In-Depth: Over 30,000+ lines of system prompt analysis
  • Timely: Continuously updated, tracking latest AI tools
  • Community-Driven: Continuously enriched through community contributions

Author/Team Introduction

system-prompts-and-models-of-ai-tools is created and maintained by x1xhlol, a community-driven open source project.

  • Author: x1xhlol
  • Background: AI tool research and reverse engineering expert
  • Project Creation Time: 2024
  • Contact:

Project Statistics

  • ⭐ GitHub Stars: 131,000
  • 🍴 Forks: 33,300
  • 📦 Version: Continuously updated (Latest update: 2026-03-08)
  • 📄 License: GPL-3.0
  • 🌐 Official Website: GitHub Repository

Project Characteristics:

  • High Popularity: 131k+ Stars, 33.3k+ Forks, highly recognized by the community
  • Active Maintenance: Project continuously updated, tracking latest AI tools
  • Open Source and Free: Uses GPL-3.0 license, free to use and modify

Main Features

Core Functionality

The core function of system-prompts-and-models-of-ai-tools is to collect and organize system prompts, internal tools, and AI model information from mainstream AI tools, providing resources for developers to learn and research:

  1. System Prompt Collection: Collects system prompts from various AI tools, revealing how tools work
  2. Internal Tool Analysis: Analyzes internal tools and APIs used by tools
  3. AI Model Information: Organizes AI models and configuration information used by tools
  4. Architecture Insights: Analyzes overall architecture design through system prompts
  5. Best Practices: Extracts best practices and design patterns from system prompts

Use Cases

  1. Learning AI Tool Design

    • Understand how tools like Cursor, Claude Code, Windsurf work
    • Learn system prompt writing techniques and best practices

  2. Researching AI Agent Architecture

    • Analyze architecture design of different AI Agents
    • Research tool selection, model configuration, prompt engineering, etc.

  3. Developing Your Own AI Tools

    • Reference existing tool system prompts to design your own tools
    • Learn best practices, avoid reinventing the wheel

  4. Education and Training

    • Used for AI tool-related education and training
    • Help beginners understand internal mechanisms of AI tools

  5. Security Research

    • Research security and privacy protection of AI tools
    • Identify potential security risks and vulnerabilities

Quick Start

Browse Resources

# Clone repository
git clone https://github.com/x1xhlol/system-prompts-and-models-of-ai-tools.git
cd system-prompts-and-models-of-ai-tools
 
# Browse different tool directories
ls -la
 
# View Cursor system prompts
cat "Cursor Prompts"/*.md
 
# View Claude Code system prompts
cat "Anthropic"/*.md

Browse Online

Visit the GitHub repository directly to browse different tool directories and files.

Core Features

  1. Wide Coverage

    • 30+ Mainstream AI Tools: Including Cursor, Claude Code, Windsurf, Devin AI, v0, etc.
    • Multiple Categories: Code assistants, AI agents, code generation, AI search, etc.
    • Continuously Updated: Tracking latest AI tools and versions

  2. High Content Depth

    • Over 30,000+ Lines: Detailed analysis of system prompts and model information
    • Structured Organization: Organized by tools for easy search and learning
    • Detailed Annotations: Includes tool descriptions, use cases, architecture analysis, etc.

  3. Clear Tool Classification

    • Code Assistants: Cursor, Claude Code, Windsurf, VSCode Agent, etc.
    • AI Agents: Devin AI, Trae, Traycer AI, etc.
    • Code Generation: v0, Lovable, Same.dev, etc.
    • AI Search: Perplexity, Poke, etc.
    • Other Tools: NotionAI, Replit, Xcode, etc.

  4. System Prompt Analysis

    • Complete Prompts: Collects complete system prompt content
    • Architecture Analysis: Analyzes overall architecture design of tools
    • Best Practices: Extracts design patterns and best practices

  5. Internal Tool Information

    • Tool Lists: Organizes internal tools and APIs used by tools
    • Tool Descriptions: Describes tool functions and usage
    • Integration Methods: Analyzes tool integration and invocation methods

  6. AI Model Information

    • Model Lists: Organizes AI models used by tools
    • Model Configuration: Analyzes model configuration and parameters
    • Performance Comparison: Compares performance and characteristics of different models

  7. Community Contributions

    • Open Contributions: Welcomes community contributions of new tools and updates
    • Issue Tracking: Tracks issues and suggestions through Issues
    • Pull Requests: Submits contributions through PRs

  8. Security Reminders

    • Security Warnings: Reminds AI startups to pay attention to data security
    • Security Recommendations: Provides AI system security recommendations and best practices
    • Security Tools: Recommends security tools like ZeroLeaks

Project Advantages

Comparison Itemsystem-prompts-and-models-of-ai-toolsOther Resource CollectionsOfficial Documentation
Coverage✅ 30+ mainstream tools⚠️ Limited tools⚠️ Single tool
Content Depth✅ 30,000+ lines analysis⚠️ Brief introductions⚠️ Public information
System Prompts✅ Complete collection❌ None❌ Not public
Internal Tools✅ Detailed analysis❌ None⚠️ Partially public
Update Frequency✅ Continuously updated⚠️ Irregular✅ Official updates
Community Contributions✅ Open contributions⚠️ Limited❌ None

Why Choose This Project?

  • Comprehensive: Covers 30+ mainstream AI tools with comprehensive content
  • In-Depth: Over 30,000+ lines of system prompt analysis with sufficient depth
  • Learning Value: Helps developers understand AI tool principles and design thinking
  • Research Value: Provides references and inspiration for AI Agent architecture design
  • High Popularity: 131k+ Stars, 33.3k+ Forks, highly recognized by the community
  • Continuously Updated: Project continuously updated, tracking latest AI tools

Detailed Project Analysis

Project Structure

The project is organized by tools, with each tool having its own directory:

system-prompts-and-models-of-ai-tools/
├── Cursor Prompts/          # Cursor system prompts
├── Anthropic/              # Claude Code system prompts
├── Windsurf/                # Windsurf system prompts
├── Devin AI/                # Devin AI system prompts
├── v0 Prompts and Tools/    # v0 system prompts and tools
├── Trae/                    # Trae system prompts
├── Perplexity/              # Perplexity system prompts
├── Replit/                  # Replit system prompts
├── Open Source prompts/     # Open source tool system prompts
├── assets/                  # Project resource files
└── README.md                # Project documentation

Organization Method:

  • By Tool: Each tool has an independent directory
  • Structured Files: Uses Markdown format for easy reading and editing
  • Detailed Annotations: Includes tool descriptions, use cases, architecture analysis, etc.

Tool Coverage

Code Assistant Category:

  • Cursor: AI code editor
  • Claude Code: Anthropic's AI code assistant
  • Windsurf: AI code editor
  • VSCode Agent: VSCode's AI agent
  • Xcode: Apple's AI code assistant

AI Agent Category:

  • Devin AI: Autonomous AI software engineer
  • Trae: AI code agent
  • Traycer AI: AI code tracing tool
  • Manus: AI agent tools

Code Generation Category:

  • v0: Vercel's AI UI generation tool
  • Lovable: AI app generation tool
  • Same.dev: AI code generation tool
  • Leap.new: AI app generation tool

AI Search Category:

  • Perplexity: AI search engine
  • Poke: AI search tool

Other Tools:

  • NotionAI: Notion's AI assistant
  • Replit: Online IDE's AI features
  • Orchids.app: AI app platform
  • Qoder: AI code assistant
  • Z.ai Code: AI code tool

System Prompt Analysis

Role of System Prompts:

  1. Define Role: Clarify AI tool's role and responsibilities
  2. Set Rules: Define tool's behavioral guidelines and constraints
  3. Provide Context: Provide necessary context information for AI
  4. Guide Behavior: Guide AI tool's execution flow and decisions

Characteristics of System Prompts:

  • Structured: Uses structured format to organize content
  • Detailed: Includes detailed rules and descriptions
  • Extensible: Supports dynamic extension and updates
  • Multi-Language: Supports multiple programming languages and scenarios

What We Learn from System Prompts:

  1. Architecture Design: Understand overall architecture design of tools
  2. Tool Selection: Learn how to select and combine tools
  3. Prompt Engineering: Learn best practices for prompt writing
  4. Error Handling: Understand error handling and recovery mechanisms
  5. User Experience: Understand user experience design thinking

Internal Tool Analysis

Tool Categories:

  1. Code Operation Tools: File read/write, code editing, Git operations, etc.
  2. Search Tools: Code search, documentation search, web search, etc.
  3. Analysis Tools: Code analysis, dependency analysis, performance analysis, etc.
  4. Generation Tools: Code generation, documentation generation, test generation, etc.
  5. Integration Tools: API calls, database operations, cloud service integration, etc.

Tool Design Patterns:

  • Tool Chain Pattern: Multiple tools combined to complete complex tasks
  • Pipeline Pattern: Tools execute sequentially, output of previous tool as input of next tool
  • Parallel Pattern: Multiple tools execute in parallel to improve efficiency
  • Conditional Pattern: Select different tools based on conditions

AI Model Information

Model Types:

  1. Code Models: Models specifically for code understanding and generation
  2. General Models: General models that can handle multiple tasks
  3. Multimodal Models: Support text, code, images, and other inputs

Model Configuration:

  • Temperature Parameter: Controls randomness of output
  • Max Tokens: Limits output length
  • Stop Words: Defines conditions for stopping generation
  • System Prompts: Defines model behavior and role

Learning Value

Value for Developers:

  1. Understand Tool Principles: Deeply understand how AI tools work
  2. Learn Best Practices: Learn best practices from system prompts
  3. Design Reference: Provide references for your own AI tool design
  4. Avoid Duplication: Avoid reinventing the wheel, reuse existing designs

Value for Researchers:

  1. Architecture Research: Research architecture design of different AI Agents
  2. Prompt Engineering: Research best practices for prompt engineering
  3. Tool Selection: Research strategies for tool selection and combination
  4. Performance Optimization: Research methods and techniques for performance optimization

Security Considerations

Security Warnings:

Project maintainers remind AI startups to pay attention to data security:

  • System Prompt Leaks: Exposed system prompts may be maliciously exploited
  • Internal Tool Leaks: Leaked internal tool information may be abused
  • Model Configuration Leaks: Leaked model configurations may be reverse engineered

Security Recommendations:

  1. Access Control: Limit access permissions to system prompts and internal tools
  2. Encrypted Storage: Encrypt sensitive information storage
  3. Audit Logs: Record access and operation logs
  4. Regular Updates: Regularly update system prompts and tool configurations

Project Address and Resources

Official Resources

Related Resources

  • ZeroLeaks: AI system security audit service
  • Latitude: LLM production predictability platform
  • DeepWiki: Codebase documentation generation tool

Supporting the Project

Project maintainers provide multiple support methods:

Target Audience

  • AI Tool Developers: Need to understand how existing tools work and their design thinking
  • AI Agent Researchers: Need to research AI Agent architecture design and best practices
  • Prompt Engineers: Need to learn best practices for prompt writing
  • Educators: Need to use for AI tool-related education and training
  • Security Researchers: Need to research security and privacy protection of AI tools

Visit my personal homepage to find more useful knowledge and interesting products

← Back to Blog