About Pdf Toolbox MCP
Pdf Toolbox MCP is an MCP server in the Security category: local-first PDF processing MCP server — OCR write-back, page surgery, encryption (OCRmyPDF + Poppler + qpdf).
Install
git clone https://github.com/twoer/pdf-toolbox-mcpThis server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.
Tools (25)
batch_ocrWhole-directory OCR with per-file results, retries, timeouts
Parameters not published.
check_repairStructural check; `repair=true` rebuilds damaged files
Parameters not published.
compress_pdfCompress, optionally down a quality ladder until hitting `target_mb`
Parameters not published.
dependency_statusProbe system tools + install commands
Parameters not published.
doctorOne-shot onboarding check: imports, dependency probe, README paths
Parameters not published.
edit_metadataSet/clear Title/Author/… (docinfo + XMP)
Parameters not published.
extract_attachmentsPull embedded attachment files
Parameters not published.
extract_imagesPull embedded images (inventory or PNG files)
Parameters not published.
extract_textLayout-aware text, exact page ranges `1-3,5`, per-page mode
Parameters not published.
fill_formFill AcroForm fields (missing fields reported)
Parameters not published.
Show all 25 tools
is_searchableSmart routing: text density check → recommends `extract_text` or `ocr_pdf`
Parameters not published.
linearizeWeb-optimized progressive-loading output
Parameters not published.
list_fontsFont audit — non-embedded fonts risk missing glyphs on other machines
Parameters not published.
locate_textFind where text occurs: page + bounding boxes (PDF points, top-left origin) — the foundation for redaction & highlighting
Parameters not published.
merge_pdfsOrdered merge
Parameters not published.
ocr_pdf**OCR write-back**: scan → searchable PDF (deskew, skip/redo, lang fallback)
Parameters not published.
pdf_infoPages, encryption status, metadata — always call first
Parameters not published.
protect_pdfAES-256 + granular permissions (print/extract/modify/…)
Parameters not published.
redact**True redaction**: affected pages rasterized + opaque boxes — redacted text physically unrecoverable, other pages keep their text layer (`rasterize_all=true` for max protection)
Parameters not published.
redact_textRedact **by content**: locate every occurrence of the given keywords and black them out — no manual coordinates needed
Parameters not published.
render_pagesPNG per page, `return_images=true` streams image blocks to the vision model
Parameters not published.
rotate_pages90/180/270 on selected pages
Parameters not published.
sanitizePublishing hygiene: strip JS/OpenAction/metadata/attachments
Parameters not published.
split_pdfBy ranges or every N pages
Parameters not published.
unlock_pdfDecrypt with **user password**, output a clean decrypted file; password sent via stdin
Parameters not published.
Ask AI
Ask AI about Pdf Toolbox MCP
Powered by Claude · Grounded in docs
Security checks
6 checks have not been run yet.