Pass Markdown directly instead of HTML

[DidacGit]

Hello,

Is there a way for to directly pass markdown instead of HTML to the request?
What I want to do is parse a PDF -> MD -> LLM Request.
Here's how I tried to do it (following Crawl4AI Docs - Local Files):

    # Download PDF and generate MD:
    response = requests.get(url)
    response.raise_for_status()
    content = response.content
    buffer = io.BytesIO(content)
    doc = open(stream=buffer, filetype="pdf")
    md_text = pymupdf4llm.to_markdown(doc, show_progress=False)
    # Do an LLM request:
    url_to_request = f"raw:{md_text}"
    # ... CrawlerRunConfig with LLMExtractionStrategy etc.

But the result is kinda weird:

• result.cleaned_html: I see it added some markup elements (e.g. <p>)
• result.html: It is the exact same as md_text
• result.markdown and result.markdown_v2.raw_markdown: It is like md_text but in a single line, for some reason.

In any case, the result.extracted_content seems to be correct! But I'm still asking just in case I'm doing something wrong. Maybe I should set LLMExtractionStrategy(input_format="") to html instead of the default markdown.

Thanks a lot in advance!