🔐 APIxAuth Protocol

Secure AI API authentication, rate limiting, and cost control - Zero infrastructure, maximum security!

🚀 Features

• 🔒 5-Layer Security - Clone keys, device fingerprinting, encryption, request signing, domain whitelisting
• ⚡ Zero Infrastructure - No servers, no databases by default (optional encrypted storage)
• 🤖 Multi-Provider - Gemini, OpenAI, Claude with auto-fallback
• 💰 Cost Control - Real-time tracking, budget limits, automatic downgrade
• 📊 Rate Limiting - Smart quota management with ML-based learning
• 🔍 Built-in Search - Google Search (Gemini), Web browsing (GPT-4o)
• 🧠 Deep Thinking - Reasoning modes (o1/o3/o4, Gemini thinking)
• 📝 Audit Logging - Cryptographically signed request logs
• 🎯 39 AI Models - Latest models from all providers (Feb 2026)

📦 Installation

npm install @apixauth/protocol

Optional Dependencies (Auto-updates)

# For always-updated AI models
npm install @google/generative-ai openai @anthropic-ai/sdk

# For database storage
npm install pg mysql2 mongodb sqlite3

🎯 Quick Start

import { APIxAuth } from '@apixauth/protocol';

const api = new APIxAuth({
  providers: {
    gemini: {
      key: process.env.GEMINI_API_KEY,
      dailyLimit: 100
    },
    openai: {
      key: process.env.OPENAI_API_KEY,
      dailyLimit: 200
    }
  },
  budget: {
    daily: 10.00,
    perRequest: 1.00
  }
});

// Basic chat
const response = await api.chat('Hello!');
console.log(response.text);

// Chat with search
const searchResponse = await api.chatWithSearch('Latest AI news 2026?');

// Chat with thinking
const thinkingResponse = await api.chatWithThinking('Solve: 2x + 5 = 15');

🔐 Key Vault Security

Protect your API keys with 5-layer security:

const api = new APIxAuth({
  keyVault: {
    enabled: true,
    masterPassword: process.env.MASTER_PASSWORD,
    allowedDomains: ['myapp.com']
  },
  providers: {
    gemini: { key: 'AIzaSy...' }  // Real key
  }
});

// After initialization, real key is encrypted
// Code now contains: axk_1a2b3c4d... (clone key)
// Clone key is useless if stolen - device-locked!

Security Layers:

1. AES-256-GCM Encryption - Military-grade
2. Clone Keys - Safe references (axk_*)
3. Device Fingerprinting - Hardware-locked
4. Request Signing - Timestamp + nonce
5. Domain Whitelisting - Authorized domains only

Learn more about Key Vault →

💡 Use Cases

Express.js API

import express from 'express';
import { APIxAuth } from '@apixauth/protocol';

const app = express();
const api = new APIxAuth({ /* config */ });

app.post('/api/chat', async (req, res) => {
  const response = await api.chat(req.body.message);
  res.json(response);
});

SaaS Application

// Per-user rate limiting
const api = new APIxAuth({
  providers: {
    gemini: { key: user.apiKey, dailyLimit: user.plan.limit }
  }
});

Batch Processing

const messages = ['Question 1', 'Question 2', 'Question 3'];
const responses = await Promise.all(
  messages.map(msg => api.chat(msg))
);

More examples →

📊 Supported Models (Feb 2026)

Gemini (17 models)

• gemini-2.5-flash (default) - Fast & efficient
• gemini-2.5-pro - Most capable
• gemini-2.0-flash-thinking-exp - Deep reasoning
• Image: gemini-2.5-flash-image, imagen-4.0
• Video: veo-3.1, veo-3.0

OpenAI (11 models)

• gpt-4o (default) - Latest GPT-4
• gpt-4.5, gpt-4.5-pro - Most advanced
• o4-mini, o3, o3-pro - Reasoning models
• gpt-4.1, gpt-4.1-mini - Coding focused

Claude (11 models)

• claude-sonnet-4.5 (default) - Balanced
• claude-opus-4.5 - Most capable
• claude-haiku-4.5 - Fastest
• claude-4.1 series

Full model list →

🎨 Features

Built-in Search

// Google Search grounding (Gemini)
const response = await api.chatWithSearch('Latest tech news?', {
  provider: 'gemini'
});

// Web browsing (GPT-4o)
const response = await api.chatWithSearch('Current Bitcoin price?', {
  provider: 'openai'
});

Deep Thinking Mode

// Gemini thinking mode
const response = await api.chatWithThinking('Complex math problem', {
  provider: 'gemini'
});

// OpenAI o1/o3/o4 reasoning
const response = await api.chatWithThinking('Logic puzzle', {
  provider: 'openai',
  model: 'o4-mini'
});

console.log(response.thinkingContent); // See reasoning process

Cost Tracking

// Get usage stats
const usage = await api.getUsage();
console.log(usage.cost); // Total cost
console.log(usage.quotas); // Per-provider quotas

// Get total cost
const cost = await api.getCost();
console.log(`Spent: $${cost.toFixed(2)}`);

Database Storage

const api = new APIxAuth({
  storage: {
    type: 'database',
    options: {
      type: 'postgresql',
      connection: {
        host: 'localhost',
        database: 'apixauth',
        user: 'user',
        password: 'pass'
      },
      encryptionKey: process.env.ENCRYPTION_KEY
    }
  }
});

📚 Documentation

• Quick Start Guide
• Key Vault Explained
• Integration Guide
• Search Feature
• Thinking Mode
• All Models (2026)
• API Reference
• Architecture

🛠️ Development

# Install dependencies
npm install

# Build
npm run build

# Run tests
npm test

# Lint
npm run lint

🤝 Contributing

Contributions are welcome! Please read CONTRIBUTING.md first.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📝 License

MIT © APIxAuth

🌟 Star History

If you find this project useful, please consider giving it a star! ⭐

📧 Support

• 📖 Documentation
• 🐛 Issue Tracker
• 💬 Discussions

🎯 Roadmap

• [ ] Next.js support (browser + edge runtime)
• [ ] Streaming responses
• [ ] Function calling support
• [ ] Multi-modal inputs (images, audio)
• [ ] Agent workflows
• [ ] More AI providers

Made with ❤️ by the APIxAuth team