graph TD %%{init: {'theme': 'mc','layout': 'elk'}}%% GDriveFilesComponent-vtobb[<div><img src="/_astro/google_drive.wKmDsV2c.svg" style="height: 20px !important;width: 20px !important"/></div>Google Drive File Manager] style GDriveFilesComponent-vtobb stroke:#a170ff ParseData-orp1r[<div><img src="/_astro/braces.Djq0PW4_.svg" style="height: 20px !important;width: 20px !important"/></div>Parse Data] style ParseData-orp1r stroke:#a170ff LanggraphReactAgent-crvmw[Agent] style LanggraphReactAgent-crvmw stroke:#a170ff ChatOutput-iwxt1[<div><img src="/_astro/messages-square.BaSDmT6g.svg" style="height: 20px !important;width: 20px !important"/></div>Chat Output] style ChatOutput-iwxt1 stroke:#a170ff OpenAIModel-7glfi[<div><img src="/_astro/openAI.BhmuxEs3.svg" style="height: 20px !important;width: 20px !important"/></div>OpenAI] style OpenAIModel-7glfi stroke:#a170ff GDriveFilesComponent-vtobb -.- ParseData-orp1r linkStyle 0 stroke:#a170ff ParseData-orp1r -.- LanggraphReactAgent-crvmw linkStyle 1 stroke:#a170ff OpenAIModel-7glfi -.- LanggraphReactAgent-crvmw linkStyle 2 stroke:#a170ff LanggraphReactAgent-crvmw -.- ChatOutput-iwxt1 linkStyle 3 stroke:#a170ff

Create Questions and Answers from Documents

🧩 Overview

This workflow extracts text from a document stored on Google Drive, converts the extracted content into plain text, and then uses an OpenAI language model to generate a set of relevant questions and answers based on that content. The process automates the creation of customized quizzes, saving time for educators and content creators.

⚙️ Main Features

Retrieves and decodes files from Google Drive using a simple file‑selection interface.
Parses binary or text data into human‑readable text with a customizable template.
Feeds the parsed text to a pre‑built Langgraph React agent that orchestrates the LLM.
Employs the GPT‑4o‑mini model to generate a structured Q&A set following a system prompt.
Streams the final response into a chat‑style output for immediate review.

🔄 Workflow Steps

Component Name	Role in the Workflow	Key Inputs	Key Outputs
Google Drive File Manager	Retrieves the selected document from Google Drive and provides its raw content.	File identifier or selection, operation (e.g., Get).	File content (binary or text).
Parse Data	Converts the raw file content into plain text using a user‑defined template.	Raw file content.	Formatted text.
Agent	Orchestrates the language‑model interaction, applying the system prompt and handling memory.	Formatted text (as input message).	Generated Q&A response.
OpenAI Model	Generates the answer text based on the agent’s prompt and input.	Input message (formatted text).	Text output.
Chat Output	Displays the generated Q&A set in a chat interface for the user.	Generated response message.	Visible chat message.

🧠 Notes

The workflow relies on a Google Drive API connection; the user must provide credentials that grant read access to the target file.
The OpenAI Model component uses the GPT‑4o‑mini variant; the model name can be swapped for other OpenAI models if higher‑quality output is required.
The Parse Data step can be customized with a different template or separator, but the default configuration assumes the file’s textual content is already structured.
The Agent is configured with a system prompt that instructs the model to generate questions and answers. Adjusting the prompt or the number of iterations can influence the length and depth of the output.
Memory and streaming are disabled by default; enabling these features will increase resource usage but can provide a more interactive experience.
All components operate in a stateless manner except for the agent’s short‑term memory, which is reset between workflow executions unless persistent storage is explicitly configured.