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

> ## Agent Instructions
> Flowy is a node-based AI creative platform: you generate images, video, audio, 3D and vector on an infinite Canvas, refine on the Studio timeline, and export or publish from the same project.
> Prefer the Flowy MCP server (https://mcp.tryflowy.ai/mcp) or the REST API at https://apis.tryflowy.ai/v1 for programmatic work. Install with `flowy mcp install` from the @flowy/cli package.
> Credits are workspace-scoped. Generations reserve credits on start and only deduct on success, so failed runs refund automatically.

# Flow Playground

> Run a published flow from its public page and watch the result come in.

The **Playground** is a flow's default tab at `/app/:slug`: a form on one side and a result stage on the other. Anyone with the link can open it and see what the flow does; running it requires signing in, since the run is billed in credits to whoever starts it.

## Filling in the inputs

The input panel has three views:

* **Form**. The normal way in: a field per input, matching what the flow's author marked when they [published it](/flows/publishing). Media fields accept a drag-and-drop upload, a pick from your asset library, or a pasted URL; text fields are plain text.
* **JSON**: the same inputs as an editable request body, for adjusting values directly.
* **cURL**: a copyable command that runs the flow exactly as configured, for taking it outside the browser.

If the flow's author set example inputs, the form loads with them prefilled so there's always something reasonable to run immediately.

## Running it

**Run** starts the flow and switches the result stage to a live progress view. Each node completes in turn, and any node that finishes early shows its output right away rather than waiting for the whole run. Your workspace's credit balance is shown alongside the button.

If a run fails, the stage shows the error plainly, along with anything that did complete before the failure, and a **Retry** button that re-runs it from the same stored inputs as a fresh run.

## The result

Once a run completes, its outputs appear in the result stage: **Preview** for the rendered media/text, or **JSON** for the raw output payload. Hovering a result reveals **Save to your assets**, **Download**, and (for media) opening its full asset details.

Every run gets its own permanent, shareable page at `/runs/:runId`: a focused, read-only view of that one result, separate from the flow's Playground.

<Tip>
  Every run of a flow, including ones you didn't just watch happen, is kept in its [Requests](/flows/requests) tab, so you can come back to any past result later.
</Tip>