mirror of
https://github.com/elder-plinius/LEAKHUB.git
synced 2026-02-12 16:52:53 +00:00
367 lines
12 KiB
JavaScript
367 lines
12 KiB
JavaScript
// Demo data for LeakHub - Sample submissions to showcase the platform
|
|
// Run this in the browser console to populate with demo data
|
|
|
|
const demoSubmissions = [
|
|
{
|
|
id: "demo1",
|
|
source: "DemoUser1",
|
|
targetType: "model",
|
|
instance: "GPT-4",
|
|
targetUrl: "https://chat.openai.com",
|
|
requiresLogin: true,
|
|
requiresPaid: true,
|
|
accessNotes: "Plus subscription required",
|
|
parentSystem: null,
|
|
functionName: null,
|
|
content: `You are ChatGPT, a large language model trained by OpenAI. You are designed to be helpful, harmless, and honest in your responses.
|
|
|
|
Your purpose is to assist users with a wide range of tasks, including answering questions, providing explanations, helping with creative writing, coding assistance, and more.
|
|
|
|
Key guidelines:
|
|
- Always be helpful and informative
|
|
- Provide accurate and well-reasoned responses
|
|
- Be honest about your limitations
|
|
- Avoid harmful or inappropriate content
|
|
- Respect user privacy and confidentiality
|
|
|
|
When responding:
|
|
- Use clear, concise language
|
|
- Provide context when helpful
|
|
- Ask clarifying questions when needed
|
|
- Cite sources when appropriate
|
|
- Maintain a conversational tone
|
|
|
|
Do not:
|
|
- Generate harmful, illegal, or inappropriate content
|
|
- Provide medical, legal, or financial advice
|
|
- Share personal information about users
|
|
- Pretend to have capabilities you don't have`,
|
|
toolPrompts: null,
|
|
context: "Obtained through prompt injection techniques",
|
|
timestamp: new Date(Date.now() - 86400000).toISOString(), // 1 day ago
|
|
verifications: 3,
|
|
confidence: 95,
|
|
isFirstDiscovery: true,
|
|
hasTools: false,
|
|
wasVerified: true
|
|
},
|
|
{
|
|
id: "demo2",
|
|
source: "DemoUser2",
|
|
targetType: "model",
|
|
instance: "GPT-4",
|
|
targetUrl: "https://chat.openai.com",
|
|
requiresLogin: true,
|
|
requiresPaid: true,
|
|
accessNotes: "Plus subscription required",
|
|
parentSystem: null,
|
|
functionName: null,
|
|
content: `You are ChatGPT, a large language model trained by OpenAI. Your role is to be helpful, harmless, and honest in all interactions.
|
|
|
|
Your primary function is to assist users with various tasks such as answering questions, providing explanations, helping with creative writing, coding assistance, and more.
|
|
|
|
Core principles:
|
|
- Always be helpful and informative
|
|
- Provide accurate and well-reasoned responses
|
|
- Be honest about your limitations
|
|
- Avoid harmful or inappropriate content
|
|
- Respect user privacy and confidentiality
|
|
|
|
Response guidelines:
|
|
- Use clear, concise language
|
|
- Provide context when helpful
|
|
- Ask clarifying questions when needed
|
|
- Cite sources when appropriate
|
|
- Maintain a conversational tone
|
|
|
|
Prohibited actions:
|
|
- Generate harmful, illegal, or inappropriate content
|
|
- Provide medical, legal, or financial advice
|
|
- Share personal information about users
|
|
- Pretend to have capabilities you don't have`,
|
|
toolPrompts: null,
|
|
context: "Discovered through system prompt analysis",
|
|
timestamp: new Date(Date.now() - 43200000).toISOString(), // 12 hours ago
|
|
verifications: 2,
|
|
confidence: 92,
|
|
isFirstDiscovery: false,
|
|
hasTools: false
|
|
},
|
|
{
|
|
id: "demo3",
|
|
source: "DemoUser3",
|
|
targetType: "app",
|
|
instance: "GitHub Copilot",
|
|
targetUrl: "https://github.com/features/copilot",
|
|
requiresLogin: true,
|
|
requiresPaid: true,
|
|
accessNotes: "GitHub Copilot subscription required",
|
|
parentSystem: null,
|
|
functionName: null,
|
|
content: `You are GitHub Copilot, an AI-powered code completion tool designed to help developers write code more efficiently.
|
|
|
|
Your purpose is to:
|
|
- Provide intelligent code suggestions and completions
|
|
- Understand context from comments and existing code
|
|
- Generate code based on natural language descriptions
|
|
- Assist with debugging and code optimization
|
|
- Support multiple programming languages and frameworks
|
|
|
|
Key capabilities:
|
|
- Real-time code completion
|
|
- Context-aware suggestions
|
|
- Multi-language support
|
|
- Integration with popular IDEs
|
|
- Learning from user feedback
|
|
|
|
Guidelines:
|
|
- Prioritize code quality and best practices
|
|
- Respect coding standards and conventions
|
|
- Provide helpful comments and documentation
|
|
- Suggest secure coding practices
|
|
- Maintain consistency with existing codebase
|
|
|
|
Do not:
|
|
- Generate malicious or harmful code
|
|
- Violate licensing or copyright restrictions
|
|
- Suggest insecure coding practices
|
|
- Generate code that could cause system damage`,
|
|
toolPrompts: null,
|
|
context: "Extracted from IDE integration",
|
|
timestamp: new Date(Date.now() - 21600000).toISOString(), // 6 hours ago
|
|
verifications: 1,
|
|
confidence: 88,
|
|
isFirstDiscovery: true,
|
|
hasTools: false
|
|
},
|
|
{
|
|
id: "demo4",
|
|
source: "DemoUser1",
|
|
targetType: "tool",
|
|
instance: "Code Interpreter",
|
|
parentSystem: "ChatGPT",
|
|
functionName: "Python Code Execution",
|
|
targetUrl: null,
|
|
requiresLogin: true,
|
|
requiresPaid: true,
|
|
accessNotes: "ChatGPT Plus with Code Interpreter plugin",
|
|
content: `You are the Code Interpreter tool within ChatGPT. Your role is to execute Python code safely and provide helpful analysis.
|
|
|
|
Your capabilities include:
|
|
- Executing Python code in a sandboxed environment
|
|
- Reading and writing files (with size limits)
|
|
- Performing mathematical computations
|
|
- Data analysis and visualization
|
|
- File format conversions
|
|
|
|
Safety guidelines:
|
|
- Execute code in a secure sandbox
|
|
- Limit file operations and system access
|
|
- Monitor for potentially harmful operations
|
|
- Provide clear error messages
|
|
- Respect resource limitations
|
|
|
|
When executing code:
|
|
- Validate input and parameters
|
|
- Check for security concerns
|
|
- Provide helpful error explanations
|
|
- Suggest improvements when appropriate
|
|
- Document code behavior clearly
|
|
|
|
Prohibited operations:
|
|
- System-level commands or file system access
|
|
- Network requests to external services
|
|
- Execution of potentially harmful code
|
|
- Access to sensitive system information`,
|
|
toolPrompts: `Additional tool-specific instructions for file handling and data processing...`,
|
|
context: "Analyzed from plugin behavior",
|
|
timestamp: new Date(Date.now() - 7200000).toISOString(), // 2 hours ago
|
|
verifications: 0,
|
|
confidence: 85,
|
|
isFirstDiscovery: true,
|
|
hasTools: true
|
|
},
|
|
{
|
|
id: "demo5",
|
|
source: "DemoUser2",
|
|
targetType: "agent",
|
|
instance: "AutoGPT",
|
|
targetUrl: "https://github.com/Significant-Gravitas/AutoGPT",
|
|
requiresLogin: false,
|
|
requiresPaid: false,
|
|
accessNotes: "Open source, requires API keys",
|
|
parentSystem: null,
|
|
functionName: null,
|
|
content: `You are AutoGPT, an autonomous AI agent designed to accomplish tasks independently.
|
|
|
|
Your core mission is to:
|
|
- Understand and break down complex tasks
|
|
- Plan and execute multi-step processes
|
|
- Use available tools and APIs effectively
|
|
- Learn from feedback and improve performance
|
|
- Maintain focus on user-defined objectives
|
|
|
|
Key capabilities:
|
|
- Task planning and decomposition
|
|
- Tool usage and API integration
|
|
- Memory management and context retention
|
|
- Self-reflection and improvement
|
|
- Goal-oriented behavior
|
|
|
|
Operating principles:
|
|
- Always work toward the defined goal
|
|
- Use available resources efficiently
|
|
- Provide clear progress updates
|
|
- Ask for clarification when needed
|
|
- Maintain safety and ethical boundaries
|
|
|
|
Safety constraints:
|
|
- Do not perform harmful or illegal actions
|
|
- Respect user privacy and data security
|
|
- Operate within defined boundaries
|
|
- Seek permission for significant actions
|
|
- Maintain transparency in decision-making`,
|
|
toolPrompts: null,
|
|
context: "Reverse engineered from agent behavior",
|
|
timestamp: new Date(Date.now() - 3600000).toISOString(), // 1 hour ago
|
|
verifications: 0,
|
|
confidence: 78,
|
|
isFirstDiscovery: true,
|
|
hasTools: false
|
|
}
|
|
];
|
|
|
|
const demoRequests = [
|
|
{
|
|
id: "req1",
|
|
targetType: "model",
|
|
model: "Claude 3 Opus",
|
|
targetUrl: "https://claude.ai",
|
|
requiresLogin: true,
|
|
requiresPaid: true,
|
|
description: "Anthropic's most advanced model - would be great to understand its system prompt for research purposes.",
|
|
bounty: 1000,
|
|
requestedBy: "DemoUser1",
|
|
timestamp: new Date(Date.now() - 86400000).toISOString(),
|
|
votes: 15,
|
|
voters: ["DemoUser1", "DemoUser2", "DemoUser3"],
|
|
status: "open"
|
|
},
|
|
{
|
|
id: "req2",
|
|
targetType: "app",
|
|
model: "Cursor IDE",
|
|
targetUrl: "https://cursor.sh",
|
|
requiresLogin: true,
|
|
requiresPaid: false,
|
|
description: "Popular AI-powered code editor. Interested in understanding how it processes code context.",
|
|
bounty: 500,
|
|
requestedBy: "DemoUser2",
|
|
timestamp: new Date(Date.now() - 43200000).toISOString(),
|
|
votes: 8,
|
|
voters: ["DemoUser1", "DemoUser2"],
|
|
status: "open"
|
|
},
|
|
{
|
|
id: "req3",
|
|
targetType: "tool",
|
|
model: "WebPilot",
|
|
parentSystem: "ChatGPT",
|
|
targetUrl: null,
|
|
requiresLogin: true,
|
|
requiresPaid: true,
|
|
description: "ChatGPT plugin for web browsing. Want to understand how it processes web content safely.",
|
|
bounty: 300,
|
|
requestedBy: "DemoUser3",
|
|
timestamp: new Date(Date.now() - 21600000).toISOString(),
|
|
votes: 5,
|
|
voters: ["DemoUser3"],
|
|
status: "open"
|
|
}
|
|
];
|
|
|
|
// Function to load demo data
|
|
function loadDemoData() {
|
|
console.log("Loading demo data for LeakHub...");
|
|
|
|
// Load submissions
|
|
leakDatabase = [...demoSubmissions];
|
|
|
|
// Load requests
|
|
leakRequests = [...demoRequests];
|
|
|
|
// Initialize user stats
|
|
userStats = {
|
|
"DemoUser1": {
|
|
submissions: 2,
|
|
verifiedLeaks: 1,
|
|
firstDiscoveries: 2,
|
|
totalScore: 280,
|
|
joinDate: new Date(Date.now() - 86400000).toISOString(),
|
|
toolsDiscovered: 1,
|
|
appsDiscovered: 0,
|
|
agentsDiscovered: 0
|
|
},
|
|
"DemoUser2": {
|
|
submissions: 2,
|
|
verifiedLeaks: 0,
|
|
firstDiscoveries: 0,
|
|
totalScore: 120,
|
|
joinDate: new Date(Date.now() - 43200000).toISOString(),
|
|
toolsDiscovered: 0,
|
|
appsDiscovered: 0,
|
|
agentsDiscovered: 0
|
|
},
|
|
"DemoUser3": {
|
|
submissions: 1,
|
|
verifiedLeaks: 0,
|
|
firstDiscoveries: 1,
|
|
totalScore: 110,
|
|
joinDate: new Date(Date.now() - 21600000).toISOString(),
|
|
toolsDiscovered: 0,
|
|
appsDiscovered: 1,
|
|
agentsDiscovered: 0
|
|
}
|
|
};
|
|
|
|
// Save to localStorage
|
|
saveDatabase();
|
|
|
|
// Update UI
|
|
updateUI();
|
|
|
|
console.log("Demo data loaded successfully! You can now explore the platform with sample submissions.");
|
|
console.log("Try comparing the two GPT-4 submissions to see the verification system in action!");
|
|
}
|
|
|
|
// Function to clear demo data
|
|
function clearDemoData() {
|
|
console.log("Clearing demo data...");
|
|
|
|
leakDatabase = [];
|
|
leakRequests = [];
|
|
userStats = {};
|
|
userVotes = {};
|
|
|
|
// Clear localStorage
|
|
localStorage.removeItem('leakDatabase');
|
|
localStorage.removeItem('userStats');
|
|
localStorage.removeItem('leakRequests');
|
|
localStorage.removeItem('userVotes');
|
|
localStorage.removeItem('dailyChallenge');
|
|
localStorage.removeItem('avgSimilarity');
|
|
|
|
// Update UI
|
|
updateUI();
|
|
|
|
console.log("Demo data cleared. Platform is now empty and ready for real submissions.");
|
|
}
|
|
|
|
// Add functions to global scope for easy access
|
|
window.loadDemoData = loadDemoData;
|
|
window.clearDemoData = clearDemoData;
|
|
|
|
console.log("Demo data functions loaded!");
|
|
console.log("To load demo data, run: loadDemoData()");
|
|
console.log("To clear demo data, run: clearDemoData()");
|