Getting Started

Get up and running with Niflheim-X in just a few minutes! This guide will walk you through installation, basic setup, and your first AI agent.

🔧 Installation

Python Installation

# Install from PyPI
pip install niflheim-x

# Or install with enterprise features
pip install "niflheim-x[enterprise]"

# Development installation
pip install "niflheim-x[dev]"

TypeScript/Node.js Installation

# Install the core package
npm install @niflheim-x/core

# Or with yarn
yarn add @niflheim-x/core

# For development
npm install --save-dev @niflheim-x/dev-tools

🚀 Lightning Fast

• 50KB total size vs LangChain's 50MB
• Instant startup - no 5-second import delays
• Minimal dependencies - only 3 core packages

🏢 Enterprise Ready

• Multi-language support - Python & TypeScript
• Production observability - metrics, tracing, logging
• Protocol standards - A2A and MCP support
• Advanced workflows - visual multi-agent orchestration

🛠️ Developer Friendly

• 5-minute setup from zero to production
• Type-safe with full IDE support
• Hot-swappable backends - memory, SQLite, vector stores
• Tool ecosystem - one-line function registration

Quick Start

Choose your language:

Python

pip install niflheim-x

import asyncio
from niflheim_x import Agent, OpenAIAdapter

async def main():
    # Create an agent
    agent = Agent(
        llm=OpenAIAdapter(api_key="your-key"),
        system_prompt="You are a helpful assistant.",
        memory_backend="dict"
    )
    
    # Add a tool
    @agent.tool
    def calculator(expression: str) -> float:
        """Evaluate mathematical expressions."""
        return eval(expression)  # Don't use eval() in production!
    
    # Chat with your agent
    response = await agent.chat("What's 25 * 4 + 10?")
    print(response.content)

if __name__ == "__main__":
    asyncio.run(main())

TypeScript

npm install @niflheim-x/core

import { Agent, OpenAIAdapter } from '@niflheim-x/core';

async function main() {
  // Create an agent
  const agent = new Agent({
    llm: new OpenAIAdapter({ apiKey: 'your-key' }),
    systemPrompt: 'You are a helpful assistant.',
    memoryBackend: 'memory'
  });
  
  // Add a tool
  agent.addTool({
    name: 'calculator',
    description: 'Evaluate mathematical expressions',
    execute: async (expression: string) => {
      return Function('"use strict"; return (' + expression + ')')();
    }
  });
  
  // Chat with your agent
  const response = await agent.chat("What's 25 * 4 + 10?");
  console.log(response.content);
}

main().catch(console.error);

Next Steps

📚 Learn the Basics

• Core Concepts - Understand agents, memory, and tools
• Memory Systems - Choose the right storage backend
• Tool Development - Build custom agent capabilities

🏗️ Build Something Real

• Multi-Agent Systems - Orchestrate agent teams
• Production Deployment - Scale to enterprise
• Observability - Monitor agent performance

🚀 Enterprise Features

• Workflow Engine - Visual agent orchestration
• Protocol Support - A2A and MCP integration
• Security & RBAC - Enterprise authentication

Performance Comparison

Metric	Niflheim-X	LangChain	BeeAI	AutoGen
Import Time	50ms	2-5s	~1s	~3s
Memory Usage	10MB	200MB	100MB	150MB
Bundle Size	50KB	50MB	15MB	25MB
Dependencies	3	50+	20+	30+
Time to Production	5 min	Days	Hours	Weeks

Why Developers Choose Niflheim-X

"Migrated from LangChain in 2 hours. Same functionality, 10x faster startup."
- Senior Developer at AI Startup

"Finally, an agent framework that doesn't break with every update."
- DevOps Engineer

"The TypeScript support is phenomenal. Best DX I've experienced."
- Full-Stack Developer

Community & Support

• 💬 Discord Community - Get help and share ideas
• 📝 GitHub Issues - Report bugs and request features
• 🐦 Twitter Updates - Latest news and tips
• 📖 Blog - Deep dives and tutorials

Ready to build lightning-fast AI agents? Install Niflheim-X and start building in 5 minutes!