Documentation Index
Fetch the complete documentation index at: https://docs.ag-kit.dev/llms.txt
Use this file to discover all available pages before exploring further.
Agents Overview
AG-Kit provides two main agent types for building AI applications.
Agent Types
Core Agent
Built-in agent with memory, tools, and event system
LangGraph Agent
LangGraph workflow integration with state management
Core Features
Core Agent
Complete agent implementation with tools, control flow, and streaming
LangGraph Agent
LangGraph workflow integration with state management
Quick Start
import { Agent, OpenAIProvider } from '@ag-kit/agents';
const modelProvider = new OpenAIProvider({
apiKey: process.env.OPENAI_API_KEY,
defaultModel: 'gpt-4'
});
// Define a tool with parameters
const weatherTool = {
name: 'get_weather',
description: 'Get current weather for a location',
schema: z.object({
location: z.string().describe('City name or coordinates'),
unit: z.enum(['celsius', 'fahrenheit']).default('celsius')
}),
handler: async (input: { location: string; unit: string }) => {
// Tool implementation
return { temperature: 22, condition: 'sunny', unit: input.unit };
}
};
const agent = new Agent({
name: 'my-agent',
model: modelProvider,
instructions: 'You are a helpful assistant with access to weather information.',
tools: [weatherTool]
});
const result = await agent.run('What is the weather in New York?');
console.log(result.data);
Installation
npm install @ag-kit/agents