Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.

Prefect Community

<@ULVA73B9P> Will the new UI support custom Prefect flow states/results?
For example, I’d like to have a custom `Skipped` state with a purple label. It shouldn’t be treated as either `Completed` or `Failed`, but as a separate state.
The use case is when the flow executes successfully, but doesn’t actually make any changes. In that case, the flow technically completed without errors, but `Completed` is misleading because nothing was changed. I’d like to be able to represent this with a distinct `Skipped` state in the UI.

:white_check_mark: thought for 134.4 seconds

Short answer: *partially*.

In Prefect 3.x, you *can* set a custom *state name* like `Skipped`, but you *cannot* create a brand-new underlying *state type* with its own separate UI semantics/color bucket.

So for your example:

```python
from prefect import flow
from prefect.states import Completed

@flow
def my_flow():
    return Completed(
        name="Skipped",
        message="No changes were needed"
    )
```

That would show a state named `Skipped`, but it would still be a `COMPLETED`-type state under the hood.

*What that means in practice*
- It can display as `Skipped`
- It will still be treated like `Completed` by orchestration logic
- It will *not* be a distinct third category separate from completed/failed
- You should *not* expect a custom purple label unless/until the UI explicitly supports color-by-name rather than color-by-type

Prefect’s docs make this distinction explicit: state *types* drive orchestration logic, while state *names* are mostly for display/bookkeeping:
- <https://docs.prefect.io/v3/concepts/states|States docs>

Relevant note on custom names and emitted events:
- <https://docs.prefect.io/v3/advanced/detect-zombie-flows#custom-state-names|Custom state names in docs>

So for your use case, the best current pattern is probably:

```python
from prefect import flow
from prefect.states import Completed

@flow
def sync_flow():
    changed = False

    if not changed:
        return Completed(
            name="Skipped",
            message="Execution succeeded, but no changes were applied."
        )

    return Completed(
        name="Completed",
        message="Changes were applied successfully."
    )
```

If you want, I can also show:
1. how this appears in automations/events, or
2. a cleaner pattern for distinguishing `Skipped` vs `Changed` in downstream monitoring.