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

# AI SDK

> Context your agent can understand

Agents are extremely capable of performing tasks, but they are limited by the amount of information available in their context. The Archil AI SDK integration gives agents the ability to store and retrieve information from an infinite filesystem, giving them access to the entirety of your data.

It exposes an interface they are already extremely familiar with -- the filesystem. It allows your agent to read files, write files, list directories, search with grep, and execute commands in a sandbox via [serverless execution](/compute/serverless-execution).

<Steps>
  <Step title="Create an Archil disk">
    Go to the [Archil console](https://console.archil.com) and create a new disk. Also, [create an API key](https://console.archil.com/api-keys) in the same region as your disk. Add the following environment variables to your project:

    ```sh theme={null}
    ARCHIL_API_KEY=<your-api-key>
    ARCHIL_REGION=<your-disk-region>
    ARCHIL_DISK_ID=<your-disk-id>
    ```
  </Step>

  <Step title="Install the SDK">
    ```bash theme={null}
    npm install disk @archildata/ai-sdk
    ```
  </Step>

  <Step title="Create tools">
    Create tools for your agent to interact with the disk.

    ```ts theme={null}
    import { generateText } from "ai";
    import * as archil from "disk";
    import { createDiskTools } from "@archildata/ai-sdk";

    const disk = await archil.getDisk(process.env.ARCHIL_DISK_ID!);
    const result = await generateText({
      model: "anthropic/claude-sonnet-4.6",
      prompt: "Read the quarterly reports and write a summary of the key findings for each one.",
      tools: createDiskTools(disk),
    });
    ```
  </Step>
</Steps>