epub-book-pipeline

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The pipeline ingests outsider-authored free text when it extracts text from an input EPUB file into source.md/chunks via extract_epub.py (reads XHTML content from the EPUB and writes its text into markdown that later steps (and any LLM chunk processors) will consume).