Cybersecurity competency is essential for securing jobs in federal agencies and industries. To ensure students are prepared for the workforce, universities should emphasize work skill readiness. This research proposes using large language models (LLMs) like Chat-GPT to extract skills from course materials and job postings. The extracted skills can then be cross-referenced with grades received to select the perfect candidate for a given task. Compared to other LLMs, Chat-GPT meets several important requirements such as speed, low cost, frequent updates, and robust APIs. The algorithm for skill extraction has salient features like segmenting longer assignments into chunks, inputting relevant categories of skills to increase the quantity and relevance of extracted skills, managing verbosity, and instructing the LLM to expand or summarize the skills. The more times a document is segmented, the more skills will be listed in the final aggregate list. However, both assignments and job postings are susceptible to tunnel vision with excessive segmentation. Furthermore, supplying stop words to the LLM is possible and can prevent unnecessary NLP pipelines. The findings suggest that there is a valuable middle ground when it comes to segmentation, and if assignments and job postings are segmented by questions and job requirements, it may be possible to extract a high number of quality skills without becoming a victim of tunnel vision. With the release of GPT4 next month, it will be possible to extract skills from recorded lectures, graphs, figures, and audio recordings such as phone interviews with prospective candidates. Overall, the results are promising, with thousands of extracted skills across hundreds of assignments and job postings, averaging 21 extracted skills per job posting and 25 extracted skills per assignment. The method provides both flexibility and comprehensiveness. For instance, we can specify that we want soft skills as well as hard skills, and the LLM provides such a curated list. LLMs have the potential to transform traditional natural language pipelines, and we have an exciting opportunity to take advantage of this technology.