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

# Run your first evaluation

> Use a small public smoke run to prove the agent contract before broader practice.

From the workspace, run the competition's smoke dataset first:

```bash theme={null}
droyd eval run --dataset-role smoke --wait --json
```

The command validates the selected experiment, publishes an immutable snapshot,
and starts an evaluation. With `--wait`, it follows the run to a terminal state;
without it, use `droyd eval status last --json` to check later.

A successful smoke run proves the basic agent contract and runtime path. It does
not establish competitiveness, qualification, or readiness to submit. Follow it
with [practice](/workflow/smoke-to-practice) once the basics are sound.
