Is it okay to crawl the Windows API documentation?

Question

I am a undergraduate research student studying AI. Recently, with the improvement of environments capable of utilizing LLM, I have been conducting research on code generation. One of the main issues in code generation is the feasibility of execution, as many LLMs often generate incorrect functions or arguments due to hallucination. I believe that by using Retrival Augment Generation with LLM, accurate code generation will be possible when leveraging specific information and parameter descriptions from the Windows API documentation as knowledge. Therefore, I have written code to crawl Windows API functions from Microsoft web page links. However, I became curious about the legality of this task, hence my inquiry.