How to parse SCORM content for chatbots and other AI tools

Rustici Generator is the easiest way to parse raw text and other data from your SCORM packages to build MCPs for your platform’s chatbots.

While the eLearning standard SCORM provides a reliable way to open a course in any LMS, inside it can be a maze of files and folders implemented in various ways across authoring tools. That said, asking your chatbot to understand these packages can be nearly impossible without extracting their contents. Rustici Generator was created not only to parse and extract text, media and interactions from SCORM packages, but also to understand the differences in how the standard is implemented across Articulate Storyline, Adobe Captivate and more.

How to parse SCORM content

1. Audit your content: Make sure you have all of the content you want to be parsed and organized in hand. It is very important to exclude content that might be proprietary or confidential, depending on where you feed the final parsed text into your LLM.
2. Connect the API: Next, connect Rustici Generator’s API to your application and configure its authentication. Once connected and authenticated, you are ready to use the application. You can interact with Generator’s API directly or through a user interface you have developed.
3. Import your content: You’ll start by importing your courses into Rustici Generator through asynchronous jobs. If configured, you can receive notifications about completed jobs as they conclude.
4. Extract structured data: After importing, you can request the parsed text from Generator’s API. The parsed text can be formatted as plain text or JSON. The JSON format will have more structure around text details, like interactions and text locations. For video content, VTT format is available as well.
5. Fuel your chatbot: With your output in hand, you can pull that into your chatbot application via API to fuel its MCP (Model Context Protocol).

How does Rustici Generator compare to parsing content with a general-purpose AI model?

While you can technically ask a ChatGPT or Claude AI Model to parse your SCORM packages, the results will be wildly inconsistent or purely non-existent. Through our research and conversations with our customers, we have discovered that even with the most elaborate prompts, general-purpose AI tools struggle to fully extract content from SCORM packages because of the variations in how authoring tools implement the standards.

Can I use Rustici Generator without integrating it into an application I’ve created or own?

Rustici Generator is a web service API that applications can integrate with, so you will need, at a minimum, development experience and a tool that leverages the API. If you don’t have either of these, we suggest working with your team or LMS provider to see what they’re using to parse your content.

While there is no user interface, we are actively exploring ways to integrate this new product with our other tools, such as Content Controller and SCORM Cloud, to make these features more accessible. If you want to be in the know of when these things happen, subscribe to our AI newsletter for the latest updates.