Intro To Prompt Chaining in FormWise

PreviousOpenAI's GPT Engines Overview NextAdvanced Prompt Chaining Guide

Last updated 2 months ago

Was this helpful?

Intro To Prompt Chaining in FormWise

Overview Prompt Chaining is one of the most powerful features in FormWise’s SmartForms. It allows you to create dynamic, multi-step AI workflows by feeding the output of one prompt into another, enabling you to build advanced tools for your users. Whether you're generating a detailed customer avatar and then creating tailored emails, or building sections of a business plan step-by-step, prompt chaining makes your AI tools smarter and more efficient.

Here’s how to use the Prompt Chaining feature within your SmartForms.

Getting Started with Prompt Chaining

You can access the Prompt Chaining feature within the SmartForms tool type.

Create a New SmartForm To begin using prompt chaining, first create a SmartForm tool inside FormWise.

Select "Prompt Chain" for Your Step Once inside the SmartForm, add a new step and select the “Prompt Chain” option. This will enable the chaining process where the output of one prompt is passed into the next.

Add Multiple Steps You need at least two prompt chain steps for the feature to work. Each step builds upon the output of the previous one.

Mapping Output Between Steps

When adding prompt instructions for your AI tool, you can map in the output from a previous step by referencing it with the @ symbol. Here’s how:

Use the @ Symbol In the second prompt step (or any following steps), hit the “@” key to bring up a list of the available outputs from earlier steps. For example, selecting @step_1 will map the output from Step 1 into your current step.
Provide Context Important: Don’t just map the previous step’s output without context. Make sure your instructions provide a clear understanding of what that step represents. For example:

❌ “Use @step_1 to generate the email copy.”

✅ “Using the customer avatar generated below, craft a targeted email that addresses their pain points. Customer Avatar: @step_1”

By giving the AI context and a description for the output being mapped from the previous step, you help it deliver more accurate and useful results.

Customizing Each Prompt Chain Step

Each prompt chain step can be customized to fit the exact needs of your workflow:

Custom Instructions: Tailor each step’s prompt instructions to guide the AI more effectively.
Use Different AI Engines: You can select a different AI engine for each prompt chain step, depending on the complexity or nature of the task.
Datasets: Use custom datasets in each step to refine the outputs.

This flexibility allows you to create highly dynamic, niche-specific tools that cater to your exact business needs.

Showing or Hiding Steps from Users

You have control over what your end users see by using the eye symbol next to each prompt chain step.

Visible Steps: If the eye is open, the user will see the step in the form as they interact with the tool.
Hidden Steps: If the eye is closed, the step will run in the background, and the user will only see the final result (not the intermediate steps).

This is helpful for creating tools where you want to simplify the user experience, only showing them the final output without overwhelming them with each process step.

PreviousOpenAI's GPT Engines Overview NextAdvanced Prompt Chaining Guide

Last updated 2 months ago

Was this helpful?

Here’s how to use the Prompt Chaining feature within your SmartForms.

Getting Started with Prompt Chaining

You can access the Prompt Chaining feature within the SmartForms tool type.

Create a New SmartForm To begin using prompt chaining, first create a SmartForm tool inside FormWise.

Select "Prompt Chain" for Your Step Once inside the SmartForm, add a new step and select the “Prompt Chain” option. This will enable the chaining process where the output of one prompt is passed into the next.

Add Multiple Steps You need at least two prompt chain steps for the feature to work. Each step builds upon the output of the previous one.

Mapping Output Between Steps

When adding prompt instructions for your AI tool, you can map in the output from a previous step by referencing it with the @ symbol. Here’s how:

Use the @ Symbol In the second prompt step (or any following steps), hit the “@” key to bring up a list of the available outputs from earlier steps. For example, selecting @step_1 will map the output from Step 1 into your current step.
Provide Context Important: Don’t just map the previous step’s output without context. Make sure your instructions provide a clear understanding of what that step represents. For example:

❌ “Use @step_1 to generate the email copy.”

✅ “Using the customer avatar generated below, craft a targeted email that addresses their pain points. Customer Avatar: @step_1”

By giving the AI context and a description for the output being mapped from the previous step, you help it deliver more accurate and useful results.

Customizing Each Prompt Chain Step

Each prompt chain step can be customized to fit the exact needs of your workflow:

Custom Instructions: Tailor each step’s prompt instructions to guide the AI more effectively.
Use Different AI Engines: You can select a different AI engine for each prompt chain step, depending on the complexity or nature of the task.
Datasets: Use custom datasets in each step to refine the outputs.

This flexibility allows you to create highly dynamic, niche-specific tools that cater to your exact business needs.

Showing or Hiding Steps from Users

You have control over what your end users see by using the eye symbol next to each prompt chain step.

Visible Steps: If the eye is open, the user will see the step in the form as they interact with the tool.
Hidden Steps: If the eye is closed, the step will run in the background, and the user will only see the final result (not the intermediate steps).

This is helpful for creating tools where you want to simplify the user experience, only showing them the final output without overwhelming them with each process step.