> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tryblend.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# User Result
> Marks a workflow output that's returned to API callers as a response field.
The **User Result** node marks an output of your workflow. When the workflow is published, whatever you connect into this node becomes one of the results shown on the runner and returned to API callers.
In the published API, each User Result node label turns into a key on the response `output` object.
## Inputs
Each User Result node has one or more **output slots** on its left edge. Connect the final content you want to return into a slot.
A fresh User Result node starts with a single empty slot. Add more with **Add new output** in the settings panel, and remove extras with **Delete output**.
| Handle | Type | Required | Notes |
| ----------- | ----- | ----------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Output slot | `any` | At least one connection | Accepts any data type — `text`, `image`, `video`, `audio`, `file`, or `any`. A slot adopts the color of whatever you connect to it. Each slot takes a single upstream connection; add more slots when one User Result should collect several pieces of final content. |
Slots accept any media or text part. They do **not** accept `tool-list` connections — those only flow between model and tool nodes.
## Outputs
The User Result node is a **terminal node** — it has no output handle on its right edge. Nothing connects downstream of it; it's the end of a branch.
## Settings
* **Result label** — names the field in the published API. "Caption" becomes `caption`; "Reference Image" becomes `reference_image`.
* **Slot labels** — each slot can show an editable display label, falling back to the upstream node's name when left blank.
* **Add new output** — adds another slot so one User Result node can collect several final outputs.
* **Delete output** — removes a slot (available once the node has more than one).
A live preview under each slot shows the content currently flowing into it.
## In the published API
Each User Result node's **label is normalized into a key** in the API `output` object. Each key maps to an **array of output items**. Each output item has `id`, `item_index`, `status`, and `parts`.
The item array is what preserves batches and list outputs. A normal one-shot chain returns one item. A batched chain returns one item per successful iteration.
Each item can still contain several parts, such as text plus a file, or several files returned by one model call.
Publishing asks you to rename User Result nodes when two labels normalize to the same API key. Use explicit labels to keep your API contract stable.
See Output parts for the full part schema, and Review and Publish for how inputs and outputs map to a published workflow.
## When to use
Add a User Result node for every distinct output you want the API to return. A workflow can have several — for example one for an `image` and one for a `caption`. Anything not connected to a User Result node stays internal and is not returned to callers.
## Connects to
Typically fed by the node that produces your final content — a model/generation node, a text node, or any node whose output is the answer you want to expose. Connect that node's output handle into a User Result slot.
## Next steps
How output items and `text`, `image`, `video`, `file`, and `json` parts are returned.
How batches and list outputs become multiple User Result items.
How nodes, handles, and connections fit together in the editor.