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

# Onboard your agent

> Configure an AI agent with the Tilebox CLI and Tilebox skills.

This tutorial gives an AI agent the default Tilebox setup for repository-based work: the Tilebox CLI and Tilebox skills. After completing it, your agent can inspect available CLI commands, run commands with structured output, and use Tilebox-specific skills for datasets, jobs, and automations.

<Steps>
  <Step title="Install the Tilebox CLI">
    Install the Tilebox CLI with the install script.

    ```bash theme={"system"}
    curl -fsSL https://cli.tilebox.com/install.sh | sh
    ```

    The CLI is the default interface for coding agents because it works in the same terminal where the agent edits files, runs tests, and applies changes.
  </Step>

  <Step title="Authenticate the CLI">
    Set `TILEBOX_API_KEY` in the environment where your agent runs. You can create an API key in the [Tilebox Console](https://console.tilebox.com/settings/api-keys).

    ```bash theme={"system"}
    export TILEBOX_API_KEY="YOUR_TILEBOX_API_KEY"
    ```

    Use a key with the smallest permissions needed for the task you want the agent to perform.
  </Step>

  <Step title="Install Tilebox skills">
    Install the Tilebox skills so your agent has task-level instructions for working with the CLI and Tilebox resources.

    ```bash theme={"system"}
    npx skills add tilebox/skills
    ```

    The skills cover CLI usage, dataset management, job monitoring, workflow authoring, and automations.
  </Step>

  <Step title="Optionally configure the Tilebox MCP server">
    Configure the Tilebox MCP server when your agent cannot easily use terminal commands, for example in a web-based chat tool. The server uses OAuth, so you do not need to add API keys or headers to the configuration.

    ```json theme={"system"}
    {
      "mcpServers": {
        "tilebox": {
          "url": "https://mcp.tilebox.com/mcp"
        }
      }
    }
    ```

    When the client connects, sign in with your Tilebox account and select the team the agent should use.
  </Step>

  <Step title="Verify the setup">
    Ask your agent to inspect its Tilebox capabilities before assigning a larger task.

    ```text Prompt theme={"system"}
    Check your Tilebox setup. Confirm whether the `tilebox` CLI is installed, whether you can inspect commands with `tilebox agent-context`, which Tilebox skills are available, and which datasets are available. Do not modify any Tilebox resources.
    ```
  </Step>
</Steps>

## Next steps

Continue with [Tilebox CLI](/agentic-development/tilebox-cli) for agent-friendly command patterns, or use [AI interfaces](/agentic-development/ai-interfaces) to decide when MCP is a better fit than terminal commands.