Client Side Pdf Redaction Tool — Permanently, On-Device, Free (2026)

Choosing the Right Tool

Prioritize tools using PDF.js or pdf-lib with WebAssembly backends — they parse and rewrite PDF structure rather than overlaying graphics. Verify the tool removes underlying text objects, images, and metadata streams (XMP, piece info, embedded files). Avoid canvas-based renderers; they rasterize pages, bloating file size and breaking text selection. Test with PDFs containing form fields, annotations, and layered content — many free tools miss these. Confirm the tool runs entirely in a Web Worker to prevent main-thread blocking on large documents. Check for CSP-compatible builds (no eval, inline scripts) if deploying in restricted environments. Ensure the library version is pinned; PDF.js updates frequently and breaking changes occur. Validate that the output passes PDF/A-2b validation if archival compliance matters.

Verifying Redaction Integrity

After redaction, open the output in a hex editor — search for original strings, SSNs, emails, or regex patterns. Use qpdf --check to confirm structural integrity. Run pdftotext -layout and diff against original text extraction; only redacted terms should disappear. Inspect /Contents streams for orphaned text operators (Tj, TJ, ') referencing removed content. Check /Annots arrays for stale markup annotations pointing to redacted regions. Validate that /Metadata stream is stripped or rewritten without author/creator fields. Test with PDFs containing incremental updates — ensure all revisions are rewritten, not just the latest. Automate verification in CI: script qpdf, pdftotext, and custom regex scans. Document the exact tool version, browser engine, and OS for reproducibility. Store hash of input and output for audit trails.