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

# Welcome to Agentbase

## The Serverless Agent Platform

Agentbase provides a fully managed platform for building and deploying production-grade AI agents. Focus on what your agents do - we handle all the infrastructure, scaling, and complexity.

<CardGroup cols={2}>
  <Card title="Build Agents" icon="hammer" href="/build/overview">
    Create powerful agents with zero infrastructure setup
  </Card>

  <Card title="Deploy to Production" icon="rocket" href="/deploy/overview">
    Launch agents with automatic scaling and 99.9% uptime
  </Card>

  <Card title="Improve & Optimize" icon="chart-line" href="/improve/traces">
    Monitor performance with traces and debugging
  </Card>

  <Card title="Agent Primitives" icon="cubes" href="/primitives/overview">
    Understand the building blocks that power your agents
  </Card>
</CardGroup>

## What Makes Agentbase Special

Every API call comes with a fully functional agent that handles tools, prompts, and capabilities automatically - no framework required.

<CardGroup cols={2}>
  <Card title="Zero Setup" icon="bolt">
    Launch your first agent in 30 seconds with `npm create agentbase@latest`
  </Card>

  <Card title="Anti-Framework" icon="ban">
    No frameworks to learn - Agentbase provides everything you need out of the box
  </Card>

  <Card title="Persistent Environments" icon="server">
    Each agent gets a full Linux environment with browser, file system, and tools
  </Card>

  <Card title="Infinite Scale" icon="infinity">
    Automatic scaling from 1 to 1000s of concurrent agents
  </Card>

  <Card title="Simple Pricing" icon="dollar-sign">
    Pay per action, not per token - predictable costs that scale with value
  </Card>

  <Card title="Production Ready" icon="shield-check">
    Enterprise-grade reliability with built-in security and monitoring
  </Card>
</CardGroup>

## Quick Start

Get your first agent running in 2 minutes:

<CodeGroup>
  ```bash Install & Run theme={null}
  # Create your agent project
  npm create agentbase@latest

  # Navigate and start
  cd my-agent
  npm run dev
  ```

  ```typescript Use in Code theme={null}
  import Agentbase from "@agentbase/sdk";

  const agentbase = new Agentbase({
    apiKey: process.env.AGENTBASE_API_KEY
  });

  const result = await agentbase.runAgent({
    message: "Analyze data.csv and create a summary report",
    mode: "base"
  });

  console.log(result.content);
  ```

  ```python Use in Python theme={null}
  from agentbase import Agentbase

  client = Agentbase(api_key=os.environ.get("AGENTBASE_API_KEY"))

  result = client.run_agent(
      message="Analyze data.csv and create a summary report",
      mode="base"
  )

  print(result.content)
  ```

  ```bash cURL theme={null}
  curl -X POST https://api.agentbase.sh/run-agent \
    -H "Authorization: Bearer YOUR_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "message": "Analyze data.csv and create a summary report",
      "mode": "base"
    }'
  ```
</CodeGroup>

<Card title="Full Quickstart Guide" icon="rocket" href="/build/quickstart">
  See the complete quickstart with examples and explanations
</Card>

## Build with Agentbase

Create powerful agents using our comprehensive primitives:

### Environment

The execution environment where agents operate:

<CardGroup cols={2}>
  <Card title="Sandbox" icon="box" href="/primitives/environment/sandbox">
    Isolated, secure execution for each agent
  </Card>

  <Card title="File System" icon="folder" href="/primitives/environment/file-system">
    Persistent file storage and management
  </Card>

  <Card title="Computer" icon="desktop" href="/primitives/environment/computer">
    Full Linux environment with shell access
  </Card>

  <Card title="Browser" icon="browser" href="/primitives/environment/browser">
    Web automation and interaction
  </Card>
</CardGroup>

### Essentials

Core capabilities that power agent functionality:

<CardGroup cols={3}>
  <Card title="Prompts" icon="message" href="/primitives/essentials/prompts">
    System prompts and instructions
  </Card>

  <Card title="Custom Tools" icon="wrench" href="/primitives/essentials/custom-tools">
    Extend with your own capabilities
  </Card>

  <Card title="Sessions" icon="messages" href="/primitives/essentials/sessions">
    Persistent conversations
  </Card>

  <Card title="Multi-Agents" icon="users" href="/primitives/essentials/multi-agents">
    Coordinate multiple agents
  </Card>

  <Card title="Parallelization" icon="chart-network" href="/primitives/essentials/parallelization">
    Concurrent execution
  </Card>

  <Card title="Self-Healing" icon="heart-pulse" href="/primitives/essentials/self-healing">
    Automatic error recovery
  </Card>
</CardGroup>

<Card title="View All Essentials" icon="list" href="/primitives/overview">
  Explore all 15 essential primitives →
</Card>

### Extensions

Advanced features for specialized use cases:

<CardGroup cols={3}>
  <Card title="Memory" icon="brain" href="/primitives/extensions/memory">
    Long-term memory and recall
  </Card>

  <Card title="RAG" icon="database" href="/primitives/extensions/rag">
    Retrieval-augmented generation
  </Card>

  <Card title="Web Search" icon="magnifying-glass" href="/primitives/extensions/web-search">
    Real-time web search
  </Card>

  <Card title="Voice" icon="microphone" href="/primitives/extensions/voice">
    Voice interaction
  </Card>

  <Card title="MCP" icon="plug" href="/primitives/extensions/mcp">
    Model Context Protocol
  </Card>

  <Card title="Workflow" icon="diagram-project" href="/primitives/extensions/workflow">
    Complex workflows
  </Card>
</CardGroup>

<Card title="View All Extensions" icon="puzzle-piece" href="/primitives/overview">
  Explore all 18 extension primitives →
</Card>

## What You Can Build

Agentbase agents handle a wide variety of tasks:

<Tabs>
  <Tab title="Development">
    **Code & Development Tasks**

    * Write, test, and debug code
    * Create REST APIs and services
    * Generate unit tests
    * Review and refactor code
    * Automate DevOps workflows

    ```typescript theme={null}
    const result = await agentbase.runAgent({
      message: "Create a REST API with authentication and user management",
      mode: "base"
    });
    ```
  </Tab>

  <Tab title="Research">
    **Research & Analysis**

    * Competitive intelligence
    * Market research
    * Data analysis and visualization
    * Report generation
    * Literature reviews

    ```typescript theme={null}
    const result = await agentbase.runAgent({
      message: "Research our top 3 competitors and create a comparison report",
      mode: "base"
    });
    ```
  </Tab>

  <Tab title="Automation">
    **Browser & Web Automation**

    * Web scraping and crawling
    * Form filling and testing
    * Price monitoring
    * Lead generation
    * Content aggregation

    ```typescript theme={null}
    const result = await agentbase.runAgent({
      message: "Monitor competitor pricing and alert if prices change",
      mode: "base"
    });
    ```
  </Tab>

  <Tab title="Data">
    **Data Processing**

    * ETL pipelines
    * Data transformation
    * CSV/JSON processing
    * Report generation
    * Database operations

    ```typescript theme={null}
    const result = await agentbase.runAgent({
      message: "Process these CSV files and generate a summary report",
      mode: "base"
    });
    ```
  </Tab>
</Tabs>

<Card title="See More Use Cases" icon="lightbulb" href="/build/use-cases">
  Explore comprehensive use cases and patterns →
</Card>

## Pricing

Simple, predictable pricing that scales with value:

| Mode      | Cost per Step | Best For                          |
| --------- | ------------- | --------------------------------- |
| **Flash** | \$0.0075      | Simple tasks, quick responses     |
| **Base**  | \$0.025       | Most production use cases         |
| **Max**   | \$0.05        | Complex reasoning, critical tasks |

<Tip>
  Sign up with your work email to get **free credits automatically**!
</Tip>

<Card title="View Full Pricing" icon="dollar-sign" href="/getting-started/pricing">
  See detailed pricing and plans →
</Card>

## Deploy to Production

Launch your agents with confidence:

<CardGroup cols={2}>
  <Card title="API Reference" icon="code" href="/deploy/api/run-agent">
    Complete API documentation with examples
  </Card>

  <Card title="TypeScript SDK" icon="js" href="/resources/sdk/typescript">
    Type-safe integration for Node.js apps
  </Card>

  <Card title="Python SDK" icon="python" href="/resources/sdk/python">
    Native Python client library
  </Card>

  <Card title="React SDK" icon="key" href="/resources/sdk/react">
    React SDK for building web applications
  </Card>
</CardGroup>

## Improve & Optimize

Monitor and enhance your agents in production:

<CardGroup cols={1}>
  <Card title="Traces" icon="route" href="/improve/traces">
    Debug with execution traces and monitoring
  </Card>
</CardGroup>

## Ready to Get Started?

<CardGroup cols={1}>
  <Card title="Try the Quickstart" icon="rocket" href="/build/quickstart">
    Get your first agent running in 2 minutes
  </Card>

  <Card title="Explore the Web App" icon="external-link" href="https://base.agentbase.sh/sign-up">
    Sign up and test agents in your browser
  </Card>
</CardGroup>

## Need Help?

<CardGroup cols={2}>
  <Card title="Join Discord" icon="discord" href="https://discord.com/invite/KFtqf7j9fs">
    Get support from our community
  </Card>

  <Card title="Contact Support" icon="headset" href="/contact-support">
    Reach out to our team
  </Card>

  <Card title="View Pricing" icon="dollar-sign" href="/deploy/pricing">
    See plans and pricing details
  </Card>

  <Card title="Read the Docs" icon="book" href="/build/overview">
    Explore comprehensive documentation
  </Card>
</CardGroup>
