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

# Set up your project

> Create a project, connect a model, and connect your data — all from the IDE

A **project** is the unit of work inside an Oxygen workspace — it holds
your agents, automations, data apps, and the `config.yml` that wires them
to data sources and models.

<Steps>
  <Step title="Create the project">
    From [app.oxygen-hq.com](https://app.oxygen-hq.com), open your workspace and
    click **+ New project** in the sidebar. Give it a name and Oxygen
    scaffolds a starter `config.yml` for you.
  </Step>

  <Step title="Connect a model">
    Open **Manage → Models** (or edit `config.yml` directly in the
    IDE) and add the LLM you want to use. The secret comes from
    [Manage → Secrets](/docs/guide/manage/secrets).

    ```yaml config.yml theme={null}
    models:
      - name: openai-4.1
        vendor: openai
        model_ref: gpt-4.1
        key_var: OPENAI_API_KEY
    ```
  </Step>

  <Step title="Connect a data source">
    Add the warehouse the agents will query. The simplest option is to
    drop a CSV or Parquet file into the project and let Oxygen's
    in-cloud DuckDB engine read it; switching to BigQuery, Snowflake,
    Postgres, or any other warehouse is a one-block change.

    ```yaml config.yml theme={null}
    databases:
      - name: local
        type: duckdb
        dataset: ./
    ```

    See [Connect your data](/docs/guide/build/connect-data) for the full
    decision tree across warehouses.
  </Step>

  <Step title="Save and proceed">
    Save the `config.yml`. Oxygen validates the file and surfaces any
    schema errors inline.
  </Step>
</Steps>

Now you're ready to [build an agent](/docs/guide/build/agents) on top of your
project.