PDF/EPUB/MOBI to TXT and Markdown Translation

Problem

Ebooks often arrive as EPUB, MOBI, or PDF files. Extracting the text is only the first step: the output still needs cleaning, chapter structure, and optional translation. Manual copying can break headings, while online models add token cost, privacy concerns, and connectivity requirements. This skill targets local conversion: turn common ebook files into editable text, generate Markdown with scripts, and translate when needed.

How It Works

• Format conversion: epub-to-txt, mobi-to-txt, and pdf-to-txt export single or batch files to TXT; MOBI uses mobi, while PDF can fall back to clawpdf and PyMuPDF.
• Scanned PDFs: pdf-to-txt-ocr uses local RapidOCR for PDFs without a text layer, such as scanned books or facsimiles.
• Markdown rendering: txt-to-md maps titles, authors, and subsections into #, ###, italics, and plain body text.
• Translation: txt-translate runs offline through Argos Translate, preserving chapter structure, supporting bilingual output, and resuming interrupted jobs.
• Batch workflows: workspace-batch-convert parses book lists, routes mixed formats, skips completed files, and emits a summary report.

Boundaries and Notes

The skill is best for extractable text. Image-only EPUBs still require exporting images and running OCR separately. Complex multi-column layouts may produce interleaved lines. Offline translation is suitable for reading and understanding, not for publication-grade copy. With special filenames, long paths, or Windows console encoding issues, verify that the output file exists and contains the expected content.