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

# Getting started

> Install the Auremi CLI and skill in your repo, then let your agent set up the model and start training.

You run one install command in your repo. After that, you ask your coding agent (Claude Code, Codex, or Cursor) for what you want, and it runs the Auremi CLI for you. Under each prompt you can open **What the agent runs** to see the commands.

<Steps>
  <Step title="Install the CLI and skill">
    From your repo's folder, run the command for your agent. It installs the `auremi` CLI from npm and adds the Auremi skill to the repo. You need Node.js 22 or newer.

    <CodeGroup>
      ```bash Claude Code theme={null}
      curl -fsSL https://auremi.ai/install.sh | AUREMI_AGENT=claude sh
      ```

      ```bash Codex theme={null}
      curl -fsSL https://auremi.ai/install.sh | AUREMI_AGENT=codex sh
      ```

      ```bash Cursor theme={null}
      curl -fsSL https://auremi.ai/install.sh | AUREMI_AGENT=cursor sh
      ```
    </CodeGroup>

    To install the CLI on its own, run `npm install -g @auremi/cli`.
  </Step>

  <Step title="Ask your agent to set up Auremi">
    <Prompt description="Set up the repo">
      Set up Auremi in this repo.
    </Prompt>

    The agent creates a starter model with an example notebook, checks it, and pushes the notebook. To sign in, it asks you to run `auremi login` yourself and approve the link in your browser. It never asks for your credentials in chat.

    <Accordion title="What the agent runs">
      ```bash theme={null}
      auremi login   # you run this one
      auremi init --name my-model
      auremi check
      auremi notebook push --message "First notebook"
      ```
    </Accordion>
  </Step>

  <Step title="Start a training run">
    <Prompt description="Train the model">
      Start a training run.
    </Prompt>

    Training needs a dataset. The CLI won't pick one for you, so the agent asks which one to use. Training runs from your latest commit on GitHub, so the agent commits and pushes first.

    <Accordion title="What the agent runs">
      ```bash theme={null}
      git add -A
      git commit -m "Starter model"
      git push
      auremi dataset list
      auremi train --dataset <datasetVersionId> --no-run
      auremi train --dataset <datasetVersionId> --wait
      ```
    </Accordion>
  </Step>
</Steps>

Training runs on a training worker or on your own cloud account. To set one up, see [Training workers](/training-workers).

For every command, see [CLI commands](/cli-commands).