A modular, high-performance framework for building AI agent systems with a focus on machine-to-machine communication.
- 🚀 High-performance task processing optimized for AI workloads
- 📊 Built-in metrics and monitoring with Prometheus integration
- 🔄 Automatic retries with configurable backoff
- 📝 Comprehensive logging and error tracking
- ⚡ Zero-copy data transfer between agents
- 🛡️ Type-safe APIs with full TypeScript support
- Node.js 16+
- npm
npm install nexus-aiimport { Agent, TaskStatus } from 'nexus-ai';
// Create an agent with custom configuration
const agent = new Agent({
maxConcurrency: 10,
retryAttempts: 3,
timeout: 30000
});
// Start the agent
await agent.start();
// Process a task
try {
await agent.processTask({
id: 'task-1',
type: 'inference',
data: {
model: 'gpt-4',
input: 'Hello, world!'
}
});
} catch (error) {
console.error('Task processing failed:', error);
}
// Get agent metrics
const metrics = agent.getMetrics();
console.log('Current metrics:', metrics);Visit our documentation site for:
- Detailed API reference
- Architecture overview
- Best practices and patterns
- Advanced configuration
- Performance tuning
- Example implementations
We welcome contributions! Please see our Contributing Guide for details.
MIT License - see LICENSE for details.