PDF/Office β Markdown Parser
Strip layout bloat (margins, footers, page numbers, XML noise) and convert PDF/Word docs into clean Markdown β with results you can review before use.
pdf.js + mammoth.jsClean, compress, and sanitize text before sending it to any model β reducing unnecessary prompt content. Savings depend on the input and model.
Showing all 7 tools
Reduce unnecessary prompt content β clean, compress, and sanitize text before sending it to any model. All processing stays in your browser.
Strip layout bloat (margins, footers, page numbers, XML noise) and convert PDF/Word docs into clean Markdown β with results you can review before use.
pdf.js + mammoth.jsPaste any web page HTML and strip <script>, <style>, <nav>, trackers & iframes β outputs only core semantic content. Review the result to check that useful content is retained.
DOMParser APIRemove repetitive quotes, commas & brackets by converting JSON to YAML or TOON notation β compare estimated sizes for your own data.
js-yaml.jsSplit massive files into token-sized parts (e.g., 2,000 tokens each) β choose paragraph, sentence, word, or hard boundaries. Ideal for RAG.
JSZip DownloadDelete all comments (#, //, /* */), trailing tabs/spaces, and blank-line bloat from Python, JS, C#, and more β review the output before using it; syntax and meaning can change.
Regex EngineApproximate token counter with an editable context limit β clip your text approximately from top or bottom before the AI throws a "context exceeded" error.
Approximate estimateSwap real names, phones, emails & cards with safe placeholders ([NAME_1], [PHONE_1]) via Regex β plus a downloadable mapping for your private reference.
Regex ScannerTry a different keyword β e.g., "pdf", "token", "json", "code".