Coursera Flashcards: Automating Video Note-Taking and Flashcard Generation

Author: Vadim Piccini
Date: March 2025
GitHub Repo: github.com/vadpiccini/coursera_flashcards

1. Introduction

Coursera Flashcards is a Chrome extension that automatically captures transcripts from Coursera video lectures, summarizes them using an OpenAI model, and generates flashcards. These flashcards can then be exported for Anki to enable efficient spaced-repetition learning.

• Problem Addressed: Manually taking notes and creating flashcards from lengthy videos can be time-consuming.
• My Solution: With a single click, Coursera Flashcards extracts the transcript, summarizes it, and creates Q&A pairs, giving you a compressed, high-value learning resource.

2. Relevance to Agentic AI

While this extension isn’t a fully autonomous agent, it demonstrates agentic AI principles by automating a multi-step pipeline:

1. Transcript Extraction: Gathers raw textual data from Coursera’s DOM.
2. LLM Summarization: Uses OpenAI’s GPT to summarize transcripts and produce question-answer pairs.
3. Task Workflow: Automatically stores, formats, and displays the final output—ready to export for Anki.

This approach showcases how LLMs can be integrated into everyday workflows—particularly in educational contexts—to save time and effort for learners.

3. Technical Implementation

Coursera Flashcards consists of three main parts: a background script, a content script, and UI pages (popup + settings). Below is a simplified overview:

flowchart LR
  A["User clicks 'Extract Notes'"] --> B["Content script extracts transcript"]
  B --> C["Background script calls OpenAI API"]
  C --> D["Summaries & flashcards returned"]
  D --> E["User exports Q&A to Anki"]

3.1 Key Code Snippets

Background Script – Calls the OpenAI API to generate flashcards:

function callOpenAI(textToProcess, callback) {
  const system_prompt = `Please summarize and create flashcards. ...`;
  
  fetch("https://api.openai.com/v1/chat/completions", {
    method: "POST",
    headers: { 
      "Content-Type": "application/json", 
      "Authorization": `Bearer ${apiKey}`
    },
    body: JSON.stringify({
      model: model,
      messages: [
        { role: "system", content: system_prompt },
        { role: "user", content: textToProcess }
      ],
      // Additional parameters
    })
  })
  .then(response => response.json())
  .then(data => callback(data))
  .catch(error => console.error("Error calling OpenAI:", error));
}

Content Script – Extracts transcript text from Coursera DOM elements:

function extractCourseraNotes() {
  const selector = ".css-4s48ix";
  const elements = document.querySelectorAll(selector);
  let allText = [];

  elements.forEach((element) => {
    const text = element.innerText.trim();
    if (text) allText.push(text);
  });

  return allText.length > 0 ? allText.join(" ") : null;
}

4. Installation & Usage

You can install Coursera Flashcards in one of two ways:

4.1 Install from the Chrome Web Store

1. Go to the Chrome Web Store link here.
2. Click Add to Chrome.
3. Grant the necessary permissions.

4.2 Install from the Repository

1. Download or Clone this repo from GitHub.
2. Go to chrome://extensions in your Chrome browser.
3. Enable Developer Mode (toggle in the top-right corner).
4. Click Load unpacked and select the project folder.

Using the Extension

1. Enter Your API Key: Open the extension’s Settings page and input your OpenAI API key (stored locally).
2. Navigate to a Coursera Video and click “Extract Notes.”
3. A new tab will display the summarized text and flashcards.
4. Export the flashcards to a .csv file and import them into Anki or another flashcard tool.

Note: Coursera Flashcards is not affiliated with Coursera, OpenAI, or Anki. Your key remains in your browser’s local storage and is used only when making requests to OpenAI.

5. Impact & Future Directions

• Efficiency: Significantly reduces the time spent on note-taking and flashcard creation.
• Customization: Users can switch models (e.g., GPT-4) for higher-quality summaries.
• Potential Enhancements:
  • Multi-platform expansion (edX, Udemy)
  • Multi-language support
  • Enhanced customization of flashcard formats

6. Conclusion

Coursera Flashcards demonstrates how agentic AI can simplify and accelerate knowledge acquisition: from raw transcript extraction to curated flashcards in one click. I invite you to explore, contribute, or simply enjoy the benefits of automated note-taking and efficient study workflows.

For questions or suggestions, please reach out via GitHub Issues or email coursera_flashcards@gmx.com

Submitted to the Agentic AI Innovation Challenge 2025.