First Flow

This guide will walk you through building your first flow in Prax, a simple pipeline that takes a user input and sends it to OpenAI’s GPT model.

---

1. Create a New Flow

• Open the Prax UI at http://localhost:3000

• Click "Create Flow"

• Give your flow a name, e.g. Basic GPT Flow

---

2. Add an Input Node

• Drag an Input node onto the canvas.

• This will serve as the starting point for any user message or external API input.

---

3. Add a Chat Model Node

• Drag in a Chat Model node (e.g., OpenAI).

• Configure it with your OpenAI API key (set in .env or UI settings).

• Set the prompt input to reference the previous node:

  handlebarsCopyEdit{{input}}

---

4. Connect Nodes

Click and drag from the output port of the Input node to the input of the Chat Model node.

---

5. Add Output Node (Optional)

To display the result:

• Drag in an Output node or a Chat Response node

• Connect it to the Chat Model node

---

6. Test the Flow

Click the Chat tab or click Run in the flow UI.

Enter a message like:

CopyEditWhat’s the capital of Japan?

You should receive a response from the model via the connected output node.

---

Tips

• You can add a Memory node to persist context between messages.

• Use Tool nodes (e.g. Web Search or Calculator) to enrich functionality.