Skip to main content

Generate Text with Cassidy Assistant

Jake Rosenthal avatar
Written by Jake Rosenthal
Updated over 3 weeks ago

How to set up and use Generate Text with Cassidy Assistant action

Uses a selected Cassidy Assistant to generate text (has context of your business and knowledge base). Follow these steps:

  1. Add a Generate Text with Cassidy Assistant action: While editing a workflow, click the plus button between existing blocks. This will open a modal where you can select from Cassidy's pre-configured actions. Choose the Generate Text with Cassidy Assistant action to add it to your workflow.

  2. Rename the action (if needed): If you want to make the action more descriptive or easier to reference as a variable later, enter a new name.

  3. Select an assistant: Choose the Cassidy Assistant you want to use for generating text.

    *Refer to the Assistants collection for more info.

  4. Write a prompt: In the "Prompt" field, enter the text that will guide the selected assistant in generating the desired output. You can reference variables from previous steps or the trigger to provide context.

    *Variables are inputs from the trigger, or outputs from previous steps

  5. Configure Structured Output Fields (optional): If you need the AI to return data in a specific format with defined fields, you can set up structured outputs. This allows you to specify exact field names, data types, and formats for consistent, parseable results. For more information, see Using Structured Output Fields.

  6. Configure Conversation Memory (optional): If you want your workflow to retain conversation history and carry context across runs, you can enable Conversation Memory. This allows follow-up messages to use earlier conversation details without starting fresh each time. For more information, see Using Conversation Memory.

  7. Update Advanced Model Settings (optional): If you want to fine-tune the AI model's behavior, you can adjust the following settings (under "Advanced Model Settings" tab):

    • Temperature: This setting controls the randomness and creativity of the generated text. A lower temperature (e.g., 0.1) will produce more predictable and focused results, while a higher temperature (e.g., 1.0) will lead to more diverse and creative outputs. We recommend starting with a temperature of 0.70 and adjusting it based on your specific requirements.

    • Include Citations: Toggle this option to control whether the assistant includes citations from its Knowledge Base in the generated text. When enabled, the assistant will provide references to source materials used in generating the response.

    • Data Analysis Mode: Enable this mode when working with structured data files like CSVs, XLSXs, PowerPoint, Word, or PDF templates. This setting allows the AI to write code on the fly to perform data-related actions, making it particularly helpful when analyzing data or making direct edits to files. For more information, see Analyzing data in Workflows.

Here's a sample of the type of output you can expect:
​

Did this answer your question?