> ## Documentation Index
> Fetch the complete documentation index at: https://docs.agentium.in/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart

> Get up and running with Agentium in minutes. Install, create your first agent, add tools, and stream responses.

# Quickstart

This guide walks you through installing Agentium and building your first agent in three steps.

***

## Installation

Install the core package and your preferred LLM provider:

<Tabs>
  <Tab title="npm">
    ```bash theme={null}
    npm install @agentium/core
    npm install openai
    ```
  </Tab>

  <Tab title="pnpm">
    ```bash theme={null}
    pnpm add @agentium/core
    pnpm add openai
    ```
  </Tab>

  <Tab title="yarn">
    ```bash theme={null}
    yarn add @agentium/core
    yarn add openai
    ```
  </Tab>
</Tabs>

For other providers, install `anthropic` or `@google/genai` instead of (or in addition to) `openai`.

Set your API key as an environment variable:

```bash theme={null}
export OPENAI_API_KEY=your-key
```

***

## Step 1: Create Your First Agent

Create a simple agent with a few lines of code:

```typescript index.ts theme={null}
import { Agent, openai } from "@agentium/core";

const agent = new Agent({
  name: "assistant",
  model: openai("gpt-4o"),
  instructions: "You are a helpful assistant.",
});

const result = await agent.run("What is TypeScript?");
console.log(result.text);
```

Run it:

```bash theme={null}
npx tsx index.ts
```

You now have a working agent that responds to user input.

***

## Step 2: Add Tools

Extend your agent with function calling. Define tools with Zod schemas for type-safe parameters:

```typescript weather-agent.ts theme={null}
import { Agent, openai, defineTool } from "@agentium/core";
import { z } from "zod";

const weatherTool = defineTool({
  name: "get_weather",
  description: "Get current weather for a city",
  parameters: z.object({
    city: z.string().describe("City name"),
  }),
  execute: async ({ city }) => `Weather in ${city}: 72°F, sunny`,
});

const agent = new Agent({
  name: "weather-bot",
  model: openai("gpt-4o"),
  instructions: "You help users check the weather.",
  tools: [weatherTool],
});

const result = await agent.run("What's the weather in Tokyo?");
console.log(result.text);
```

The agent will call `get_weather` when needed and incorporate the result into its response.

***

## Step 3: Stream Responses

For real-time output, use the streaming API:

```typescript stream-agent.ts theme={null}
import { Agent, openai } from "@agentium/core";

const agent = new Agent({
  name: "assistant",
  model: openai("gpt-4o"),
  instructions: "You are a helpful assistant.",
});

for await (const chunk of agent.stream("Tell me a joke")) {
  if (chunk.type === "text") {
    process.stdout.write(chunk.text);
  }
}
```

Streaming supports both text and tool-call chunks. Handle each `chunk.type` as needed for your use case.

***

## Next Steps

<CardGroup cols={2}>
  <Card title="Agents" icon="robot" href="/agents/overview">
    Learn about agent configuration, instructions, and advanced options.
  </Card>

  <Card title="Models" icon="cpu" href="/models/overview">
    Switch providers, use local models with Ollama, or build custom adapters.
  </Card>

  <Card title="Teams" icon="users" href="/teams/overview">
    Coordinate multiple agents with route, broadcast, and collaborate modes.
  </Card>

  <Card title="Workflows" icon="workflow" href="/workflows/overview">
    Build stateful multi-step workflows with conditions and parallel execution.
  </Card>
</CardGroup>