Upload your PDF and its scraped MD — missing spaces, headings, and formatting issues will be automatically checked and fixed.
Don't need to fix the whole PDF, just a chunk where words or sentences got merged together? Paste it here.
Check many PDF+MD pairs at once. Files are matched by name (e.g. 25659.pdf ↔ 25659.md).
For a scanned PDF whose existing MD scrape is unusably garbled (OCR noise, not just missing spaces) — builds a brand-new MD from scratch via OCR + layout detection. Slower than Check & Fix (can take several minutes to over an hour for a long document), so this runs in the background with a progress indicator instead of one long wait.
Scan the whole document's heading structure at a glance. Headings marked ⚠ look broken (garbled/truncated text) and are worth checking against the PDF.
Zoning/legal terms and proper nouns can look "wrong" to a spell-checker but be entirely correct — nothing here is changed automatically. Click a red word in the preview to fix it or mark it "not a typo" (remembered on this browser for future files too).
Scraped MD tables often lose their structure entirely. These are pulled straight from the PDF instead, with multi-row/merged headers auto-flattened into one clean header row — use these to replace the broken table sections in your MD.
⬇ Download Extracted Tables