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

# Introduction

> Flowyble is a multi-tenant AI-powered workflow automation platform. Create and run AI agents and workflow tools powered by leading LLMs.

## What is Flowyble?

Flowyble lets you build **AI agents** and **workflow tools** that automate complex tasks using large language models. Agents can reason, use tools, and hold multi-turn conversations — all managed through a simple web interface or REST API.

<Columns cols={2}>
  <Card title="Quickstart" icon="rocket" href="/quickstart">
    Create your first agent in under 5 minutes.
  </Card>

  <Card title="Core Concepts" icon="lightbulb" href="/concepts">
    Understand agents, tools, runs, and credits.
  </Card>
</Columns>

## Key Features

<Columns cols={2}>
  <Card title="AI Agents" icon="robot" href="/agents/overview">
    Autonomous agents that reason, call tools, and deliver results — with configurable LLM models, step limits, and conversation memory.
  </Card>

  <Card title="Workflow Tools" icon="wrench" href="/tools/overview">
    Multi-step automations composed of built-in functions and integrations. Run manually, via API, or triggered by webhooks.
  </Card>

  <Card title="Integrations" icon="plug" href="/integrations/overview">
    Connect external services via OAuth2, API keys, or webhooks. Use integration tools inside your agents and workflows.
  </Card>

  <Card title="REST API" icon="terminal" href="/api-reference/introduction">
    Execute agents and tools programmatically. Authenticate with API keys, poll for results, and integrate Flowyble into your stack.
  </Card>
</Columns>

## How it works

<Steps>
  <Step title="Create an Agent or Tool">
    Define behavior prompts, choose an LLM model, attach tools and knowledge bases.
  </Step>

  <Step title="Run it">
    Execute via the dashboard, chat interface, API call, or webhook trigger.
  </Step>

  <Step title="Get results">
    View step-by-step execution, token usage, and final output in real time via SignalR or by polling the API.
  </Step>
</Steps>