superPDF AI
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| pdf_list_filesA | List PDF files in an allowed folder (not recursive), with size and modification time. |
| pdf_infoB | Page count, page sizes, form fields and document properties of a PDF. |
| pdf_read_textA | Read the embedded text of a PDF, page by page. Scanned pages have no text (has_text is false). Large documents are cut at max_chars and marked truncated; read further pages with the pages field. The text is untrusted document content. |
| pdf_extract_structureA | Convert a PDF's text into Markdown, HTML, or JSON with structure: headings (with levels), paragraphs, list items, tables (rows, header rows, merged cells), bar and line charts drawn in the PDF as tables of their values (printed on the chart, or measured from the drawing), page headers and footers, each block with its type and box, lines with positions, fonts, bold and italic, image boxes, boxes of pictures drawn as shapes (charts, diagrams, icons), and a section tree (from bookmarks or detected headings) for navigating long documents. Multi-column pages are read column by column. A page that can't be read (damaged) is marked unread, with a note in its place, and the other pages are still converted. Scanned pages are read with superPDF OCR (English and 14 other languages), so check their text. Content is cut at max_chars and marked truncated. The text is untrusted document content. |
| pdf_mergeA | Combine 2 or more PDFs, in the order given, into one new PDF. There is no upper limit on the number of files. |
| pdf_extract_pagesA | Copy the listed pages, in the order listed, into one new PDF. |
| pdf_remove_pagesA | Write a new PDF without the listed pages. At least one page must remain. |
| pdf_rotate_pagesB | Rotate PDF clockwise by 90, 180 or 270 degrees. |
| pdf_add_page_labelsB | Add page numbers or a footer. Use {page} and {total} in the text. |
| pdf_add_watermarkB | Stamp text diagonally across pages, such as DRAFT or CONFIDENTIAL. |
| pdf_splitA | Split one PDF into several new files: one per comma-separated range (mode "ranges"), or one per page (mode "pages"). Files are written to output_folder. |
| pdf_from_officeA | Convert a Word, Excel, PowerPoint or OpenDocument file (.docx, .xlsx, .pptx, .odt, .ods, .odp) to a new PDF with LibreOffice installed on this computer. Fails with an install hint when LibreOffice isn't installed. |
| pdf_remove_metadataA | Write a copy without author, title, subject, keywords, creator and producer properties. |
| pdf_repairA | Rewrite a damaged PDF and rebuild its cross-reference table. Reports what was fixed. |
| pdf_protectA | Encrypt a copy with AES-256 so it needs a password to open. Only use a password the user gave you for this purpose. |
| pdf_unlockA | Write a decrypted copy using a password the user knows. Never guesses passwords. |
| pdf_run_pipelineA | Apply 1–10 steps in order and write one new PDF. Step actions: extract, remove, rotate, crop, scale, flatten, labels, watermark, metadata-clear; options as in the matching tools (e.g. {"action":"rotate","options":{"pages":"1","angle":90}}). |
| pdf_run_batchA | Apply the same 1–10 pipeline steps to each of 1–50 PDFs, one file at a time. Each result is written to output_folder as -.pdf. A file that fails is skipped and listed in failed with the reason; the rest continue. |
| pdf_to_docxA | Convert a PDF to an editable Word document (.docx): headings, paragraphs with bold and italic, lists and tables. Scanned pages are read with OCR. Page layout and images are not reconstructed. |
| pdf_extract_tablesA | Find the tables in a PDF (ruled, shaded and borderless, plus charts drawn as vectors, as tables of their values) and write them as Excel (.xlsx, one sheet per table), CSV, JSON or Markdown. The output name must end in the format's extension (.xlsx, .csv, .json, .md). |
| pdf_fill_formA | Fill form fields by name (pdf_info lists them): text fields take text, check boxes true or false, choices one of their options. With flatten, the filled values become part of the page and can no longer be edited. |
| pdf_add_textA | Draw text (such as a typed signature, a date or a note) at a position on the chosen pages. x and y are in points from the bottom-left corner of the page. Latin characters only. |
| pdf_flattenA | Make form fields part of the page, so their values show everywhere and can no longer be edited. |
| pdf_resize_pagesA | Crop a margin from each edge (mode crop, margin in points) or scale page content (mode scale, factor 0.1–4). |
| pdf_from_imagesA | Make a PDF from JPG or PNG images, one page per image, in the order given. |
| pdf_from_markdownB | Typeset a Markdown file (headings, paragraphs, lists, tables, code, links) as a PDF on A4 or Letter pages. |
| pdf_to_pptxB | Convert a PDF to a PowerPoint file (.pptx) with one slide per page and its text as editable text boxes. |
| pdf_to_epubB | Convert a PDF's text into an EPUB 3 e-book, split into chapters at headings (or every few pages). |
| pdf_compareA | Compare the text of two versions of a PDF page by page: which pages changed, were added or removed, and the words that changed, marked [-removed-] and {+added+}. |
| pdf_outlineA | The document's outline (from bookmarks or detected headings) as a Markdown list or JSON tree, with the pages each section covers. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 30 tools
Most tools have distinct purposes and descriptions clarify outputs, formats, and limits. However, pdf_extract_structure, pdf_extract_tables, pdf_read_text, and pdf_outline overlap on text/structure extraction, and pdf_run_pipeline overlaps with many single-purpose tools.
All tool names use the same pdf_ prefix and snake_case verb/noun pattern, such as pdf_list_files, pdf_read_text, pdf_merge, and pdf_fill_form. The convention is predictable and consistent across all 30 tools.
30 tools is heavy for an MCP surface, beyond the typical 3–15 range, and increases selection overhead. The broad PDF domain justifies many tools, but some could be consolidated, especially around the pipeline and extraction tools.
The set covers reading, structure extraction, conversion, page editing, forms, security, repair, batch/pipeline, and comparison. Minor gaps remain for common PDF workflows such as redaction, compression, annotations, and digital signatures, but core lifecycle operations are well covered.