> For the complete documentation index, see [llms.txt](https://docs.stackai.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.stackai.com/guides-and-quickstarts/workflow-builder/first-input-to-output-workflow.md).

# First input to output workflow

A workflow connects inputs, processing steps, and outputs. Use it to define how StackAI turns a request into a result.

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FSQ1Sgy933krTyK0S4JpP%2Fuploads%2Fe0mbRuOXYf2QFUZkVIgm%2FG020.mp4?alt=media&token=2ac78248-8bc6-49b1-b7de-ccb8fea7f62e>" %}

## Before you start

You need permission to create and edit a project and access to an AI model in your organization. Prepare a short paragraph you can check manually. This first workflow does not require an API, code, files, or external connections.

## Create the project

From **Projects**, select **New Project** and choose the workflow option. In the creation dialog that offers **New Workflow**, choose it to start with a blank canvas. Enter a descriptive project name and create the project.

![New workflow project creation](https://content.gitbook.com/content/SQ1Sgy933krTyK0S4JpP/blobs/kjhnWQmXTHDsJlTr79DY/assets/g020-g020-01-create.png)

For this first workflow, you need only **Text Input**, **AI Agent** (the model step), and **Output**. Use the node sidebar’s search to find these node types and drag each from the sidebar onto the canvas. If your starting project already contains them, use those existing nodes. You can also [start from a template](https://docs.stackai.com/guides-and-quickstarts/workspace-and-discovery/create-a-project-from-a-template.md) and inspect its preview for these steps before using it.

## Connect the workflow steps

Connect **Text Input → AI Agent → Output**. Drag from a node’s outgoing connection handle to the next node’s incoming handle. Follow the connecting lines to confirm the input reaches the model and the model reaches Output.

![Input, model, and Output connections](https://content.gitbook.com/content/SQ1Sgy933krTyK0S4JpP/blobs/TYTgENOXk2KSFOfybInW/assets/g020-g020-02-canvas.png)

Connections determine the execution path. References determine which values a node receives.

## Configure the model

Open **AI Agent** and choose a model available to your organization. In **Instructions**, enter:

```
Summarize the supplied text in three short bullet points.
Use only the supplied facts. Preserve names, dates, and amounts.
If there is no text to summarize, ask for it.
```

Click inside **Prompt**, type `@` to open the reference picker, and select **Text Input** and its text value. The inserted reference should appear as a reference control, rather than ordinary text containing the node’s name. Selecting that reference lets you inspect or edit its mapping.

![Model Instructions and Text Input prompt reference](https://content.gitbook.com/content/SQ1Sgy933krTyK0S4JpP/blobs/mo0YHh5hupm16Al3QwCl/assets/g020-g020-03-prompt.png)

The reference supplies the current input each time the workflow runs.

## Configure the output and run

Open Output and review its connected source.

![Output source configuration](https://content.gitbook.com/content/SQ1Sgy933krTyK0S4JpP/blobs/zHiBWSDyd8k6jsENAzs1/assets/g020-g020-04-output.png)

Use a custom template only when you need to select particular fields or change their presentation.

Enter a short paragraph with three checkable facts in **Text Input**, save the draft, and select **Run**. Inspect the input, model result, and Output. Expect three short bullets based on the paragraph. Change a fact and run again to confirm the new value reaches both the model and Output.

![First workflow run and output](https://content.gitbook.com/content/SQ1Sgy933krTyK0S4JpP/blobs/OpLDHP9uw5I6OrrRWxnW/assets/g020-g020-05-persisted.png)

Saving the workflow does not publish an interface for end users.

## Important behavior

Connecting two nodes defines the processing path. Use the reference picker to pass a value from one node into a field in another node.

## Check the result and share it

The run should show the submitted text at the input, a response from the model, and that response in Output. If the model receives no request text, check the Prompt reference. If the model answers but Output is empty, check the output mapping.

When the result is correct, [publish a Chat Assistant](https://docs.stackai.com/guides-and-quickstarts/interfaces-and-end-user-experiences/publish-a-chat-assistant.md) so users can interact with the workflow without opening the canvas.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.stackai.com/guides-and-quickstarts/workflow-builder/first-input-to-output-workflow.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
