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

# Quickstart

> Download an example agent, add your provider keys, and talk to it locally before you deploy.

Get a working agent project on your machine and talk to it in the browser. Nothing runs in
the cloud yet. You need the CLI installed and signed in; see
[Installation & Setup](/deployments/installation).

<Steps>
  <Step title="Create a project">
    `zeroruntime quickstart` downloads a ready-made example into a new folder and runs it
    locally:

    ```bash theme={null}
    zeroruntime quickstart
    ```

    Pick an example from the list, or skip the picker by naming one:

    ```bash theme={null}
    zeroruntime quickstart --template <example-id>
    ```

    This creates a `./<example-id>/` folder:

    | File | Purpose |
    | - | - |
    | `main.py` | Your agent's entrypoint and pipeline. |
    | `requirements.txt` | Python dependencies. |
    | `.env` | Provider API keys and secrets (kept local). |
    | `config.json` | Local run metadata. |

    <Tip>
      Already have an agent project? Skip this step. It needs an entrypoint named `main.py`,
      `agent.py`, or `app.py`.
    </Tip>
  </Step>

  <Step title="Add your provider keys">
    Move into the project folder and fill in `.env` with the keys for the STT, LLM, and TTS
    providers your agent uses:

    ```bash theme={null}
    cd <example-id>
    ```

    ```bash .env theme={null}
    ZERORUNTIME_AUTH_TOKEN=...   # added for you by quickstart
    DEEPGRAM_API_KEY=...         # your provider keys
    GOOGLE_API_KEY=...
    CARTESIA_API_KEY=...
    ```

    <Warning>
      Never commit `.env` or bake keys into a Docker image. Deployed agents receive them as
      [secrets](/deployments/operate/secrets).
    </Warning>
  </Step>

  <Step title="Talk to it locally">
    Run the agent and open the **playground** in your browser:

    ```bash theme={null}
    zeroruntime run . --open
    ```

    ```text Expected output theme={null}
    Running project '<example-id>'...
    Playground ready: https://playground.zeroruntime.ai/...
    Opening the playground in your browser...
    ```

    The first run creates a virtual environment and installs `requirements.txt`. To talk in your
    terminal instead, use `zeroruntime run . --console`.
  </Step>
</Steps>

## Next: deploy it

<CardGroup cols={2}>
  <Card title="Zero Runtime Cloud" icon="cloud-arrow-up" href="/deployments/zeroruntime-cloud">
    Deploy with one command. We run it for you.
  </Card>

  <Card title="Self-Hosting" icon="server" href="/deployments/self-hosting">
    Run the worker on your own infrastructure.
  </Card>
</CardGroup>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.