io.github.theluckystrike/convert-pdf-to-word-docx
Measured conclusion: Deep-tested. L2 outcome: tools_ok.
Convert PDF pages into editable Word-ready text from your AI chat, all offline.
| Author | io.github.theluckystrike |
|---|---|
| Access | remote |
| Publisher key required | No |
| Measured state | Deep-tested |
| L1 liveness | alive |
| L1 latency | 2300 ms |
| Last measured | 2026-10-24 |
Add to your agent
Copy this into your MCP client config:
{
"mcpServers": {
"convert-pdf-to-word-docx": {
"url": "https://mcp.zovo.one/mcp/convert-pdf-to-word-docx",
"transport": "streamable-http"
}
}
}
Measured tools (15)
pdf_infoCall this tool to inspect a PDF: page count, page sizes in points/mm, paper name, metadata (title, author, producer, dates) and whether it is encrypted. Read-only, never modifies the file. Free tier: unlimited.
Parameters:
path(required) — string — Name of a PDF uploaded with pdf_upload, or one this server wrote earlier
Hints: read-only no · destructive no · idempotent no · open-world no
pdf_countPage count per file plus the total, for any number of PDFs, with how many were readable. A file that cannot be read is reported on its own row and the others still count. Read-only. Free and unlimited.
Parameters:
paths(required) — array — Names of PDFs uploaded with pdf_upload
Hints: read-only no · destructive no · idempotent no · open-world no
pdf_mergeCall this tool to join several PDFs into one, in the order given. Page sizes are kept as-is, so a merged file may have mixed sizes, and the answer says so. Inputs are never modified. Free tier: up to 5 files per merge.
Parameters:
paths(required) — array — Names of the uploaded PDFs to join, in the order they should appearout_path(required) — string — Name for the merged PDF; it comes back as a download link valid for one houroverwrite— boolean — Replace out_path if a file is already there. Default false: an existing file is never overwritten
Hints: read-only no · destructive no · idempotent no · open-world no
pdf_splitCall this tool to write one new PDF per range. Ranges are 1-based and may be open-ended: "1-3,5,7-" gives pages 1-3, page 5, and 7 to the end. The input is never modified. Free tier: files up to 30 pages.
Parameters:
path(required) — string — Name of the uploaded PDF to splitranges(required) — string — Comma-separated 1-based page ranges, e.g. "1-3,5,7-". An open-ended range runs to the last pageout_path_pattern(required) — string — Name pattern for the parts: {n} is the part number (1, 2, 3...), {range} is the range itself (e.g. 1-3), {name} is the input name. Example: {name}-{range}.pdf. Each part comes back as a download link valid for one houroverwrite— boolean — Replace existing outputs. Default false: nothing is overwritten and nothing is written at all if any target exists
Hints: read-only no · destructive no · idempotent no · open-world no
pdf_pagesCall this tool to pull selected pages into one new PDF, in the order written: "2,4-6" gives four pages. Asking for a page twice copies it twice. The input is never modified. Free tier: files up to 30 pages.
Parameters:
path(required) — string — Name of an uploaded PDF (pdf_upload), or one this server wrote earlierpages(required) — string — 1-based pages and ranges to keep, in output order, e.g. "2,4-6" or "5,1,1"out_path(required) — string — Name for the extracted PDF; it comes back as a download link valid for one houroverwrite— boolean — Replace out_path if it exists. Default false
Hints: read-only no · destructive no · idempotent no · open-world no
pdf_rotateCall this tool to turn pages by a multiple of 90 degrees, clockwise for positive. Rotation is added to whatever the page already had, for a sideways scan. Writes a new file. Free tier: files up to 30 pages.
Parameters:
path(required) — string — Name of an uploaded PDF (pdf_upload), or one this server wrote earlierdegrees(required) — integer — 90, 180, 270 or -90. Positive turns clockwise. Added to the page's existing rotationout_path(required) — string — Name for the rotated PDF; it comes back as a download link valid for one hourpages— string — Which pages to turn, e.g. "1" or "2,4-6". Omit for every pageoverwrite— boolean — Replace out_path if it exists. Default false
Hints: read-only no · destructive no · idempotent no · open-world no
pdf_stampCall this tool to draw a word such as PAID or DRAFT across the pages, in a colour and position you choose. Writes a new file; input untouched. Free tier: the PAID and DRAFT presets in their preset colours.
Parameters:
path(required) — string — Name of an uploaded PDF (pdf_upload), or one this server wrote earliertext(required) — string — What to stamp. PAID and DRAFT are presets with their own colour; any other text is Proout_path(required) — string — Name for the stamped PDF; it comes back as a download link valid for one hourcolor— string — Hex code such as #1b7f3b, or a name: red, green, blue, black, gray, orange, purple. Propages— string — Which pages to stamp, e.g. "1" or "2,4-6". Omit for every pageopacity— number — 0 to 1. Default 0.35 in the centre, 0.85 in a cornerposition— string — Where on the page. Default center, which is drawn on the 45-degree diagonal like a real stampfont_size— number — Point size. By default the stamp is sized to fit the page widthoverwrite— boolean — Replace out_path if it exists. Default false
Hints: read-only no · destructive no · idempotent no · open-world no
pdf_watermark_businessCall this tool to put your business name and VAT id in the footer of every page, from the shared profile mcp-invoice and mcp-docx write. Use it before sending a document out. Pro.
Parameters:
path(required) — string — Name of an uploaded PDF (pdf_upload), or one this server wrote earlierout_path(required) — string — Name for the footed PDF; it comes back as a download link valid for one hourpages— string — Which pages, e.g. "1". Omit for every pageoverwrite— boolean — Replace out_path if it exists. Default false
Hints: read-only no · destructive no · idempotent no · open-world no
pdf_reorderCall this tool to write a new PDF with pages in the order you give. The order must name every page exactly once, so nothing drops by accident; use pdf_pages for a subset. Pro.
Parameters:
path(required) — string — Name of an uploaded PDF (pdf_upload), or one this server wrote earlierorder(required) — array — The 1-based page numbers in their new order, e.g. [3,1,2] for a three-page file. Every page must appear exactly onceout_path(required) — string — Name for the reordered PDF; it comes back as a download link valid for one houroverwrite— boolean — Replace out_path if it exists. Default false
Hints: read-only no · destructive no · idempotent no · open-world no
pdf_textread-onlyCall this tool for best-effort text extraction from standard-font PDFs. Returns nothing for a scan (no OCR), or glyph indices for a custom-encoded font, and says which case applies. Read-only. Free tier: unlimited.
Parameters:
path(required) — string — Name of the PDF to readpages— string — Which pages, e.g. "1" or "2,4-6". Omit for every page
Hints: read-only yes · destructive no · idempotent yes · open-world no
license_statusread-onlyReport this endpoint's licence state for your token as JSON: the product, the tier free or pro, why it is not Pro, and the checkout URL. Call it to explain a free-tier refusal. No arguments, nothing changes.
Hints: read-only yes · destructive no · idempotent yes · open-world no
license_activateTurn Pro on for this connection with key, an MCPL1.<payload>.<signature> issued at checkout for this server or the bundle. Data under your token stays; a wrong or expired key changes nothing. license_status confirms it.
Parameters:
key(required) — string — License key from checkout, MCPL1.<payload>.<signature>
Hints: read-only no · destructive no · idempotent no · open-world no
pdf_uploadSend a PDF to this hosted endpoint. There is no filesystem here, so instead of a path you upload the file once with pdf_upload and then pass its name as `path` (or inside `paths`) to pdf_info, pdf_count, pdf_merge, pdf_split, pdf_pages, pdf_rotate, pdf_stamp, pdf_reorder and pdf_text. Give exactly one of pdf_base64 or url. url: fetch a public file instead of pasting base64 (recommended above about 10 KB): pasting a large base64 payload is slow to write out and can stall the turn before the call is ever sent, while a url costs one line and the fetch happens here. Uploads are kept for your token between calls; pdf_files lists them and pdf_delete_upload removes one. A file this server writes is named the same way and comes back as a download link valid for one hour. The request body cap is 256 KB, which is about 190 KB of PDF once base64-encoded; a url is fetched with a 10 second timeout, at most 3 redirects, public http(s) hosts only, and is capped at 2 MB.
Parameters:
name(required) — string — Name to refer to this PDF by: 1-64 characters of letters, digits, underscore or dash, e.g. "invoice"url— string — url: fetch a public file instead of pasting base64 (recommended above about 10 KB). Public http(s) only; private, link-local and this endpoint's own zone are refusedpdf_base64— string — The PDF file, base64-encoded
Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared
pdf_filesList the PDFs stored for your token: what you uploaded, and what this server has written, with their sizes.
Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared
pdf_delete_uploadDelete one PDF stored for your token, uploaded or generated.
Parameters:
name(required) — string
Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared
Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.