pdfforge
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| pdf_infoC | Page count, metadata, encryption, form fields, attachments, and which pages have no text layer (scans that need OCR). |
| read_textC | Text of the document (or of the pages given) in reading order, page by page. |
| render_pageC | A picture of one page, to look at layout, stamps, signatures or scans. |
| searchC | Every occurrence of a word or phrase, with its page and surrounding text. |
| mergeB | Join several PDFs into one, in order. Each source becomes a bookmark. |
| splitC | One new PDF per range, for instance '1-3, 4-8, 9-'. |
| organize_pagesB | Keep only some pages, delete some, and/or rotate some (angle 90, 180 or 270). Ranges use the original page numbers. |
| create_pdfA | Build one PDF from Word, Excel, PowerPoint, OpenDocument, images or text files (office formats need LibreOffice). |
| exportA | Convert a PDF to .docx, .txt, .html, .odt or .xlsx tables, chosen by the
extension of |
| replace_textB | Replace a word or phrase everywhere it appears, rewriting the text itself in the document's own embedded fonts (not a white box drawn on top). |
| ocrA | Make a scanned PDF searchable with Tesseract: straightens upside-down and
skewed pages and reports its confidence. |
| watermarkC | Stamp a text watermark such as CONFIDENTIAL or COPY across the pages. |
| list_form_fieldsB | Fillable fields of the form: name, type, current value, page. |
| fill_formC | Fill form fields by name. Checkboxes take 'yes'/'no'. |
| find_sensitive_dataB | Find IBANs, card numbers, national register numbers, VAT numbers, e-mails, phone numbers and birth dates. IBANs, cards and register numbers are checked with their control digits. Values are masked in the answer. |
| redactA | Remove sensitive data for good: the characters leave the file, it is not a
black box drawn on top. |
| protectC | Encrypt with AES-256: a password to open and/or restricted permissions. |
| unlockB | Remove the password from a PDF you can open. |
| extract_transactionsA | Read a bank statement or an invoice list into rows of date / label / amount.
With |
| compareC | Word-by-word and visual comparison of two versions of a document. |
| detect_tamperingA | Look for signs a document was altered after it was issued: earlier revisions kept in the file, text hidden under shapes, mismatched dates and producers. |
| compressB | Make the file smaller. Presets: 'email' (smallest), 'archivage' (balanced), 'impression' (print quality). |
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 22 tools
Most tools have clearly distinct actions and outputs: read_text vs search vs extract_transactions, and find_sensitive_data vs redact, are well differentiated. Minor overlap exists between pdf_info and list_form_fields for form-field information, and between export and extract_transactions for tabular output, but descriptions provide strong cues.
Names are almost entirely predictable snake_case and action-oriented (read_text, create_pdf, replace_text, detect_tampering, compress). Single-verb names for simple actions are readable, but pdf_info and ocr are minor deviations from a strict verb_noun pattern.
At 22 tools, the server is above the typical 3–15 range, but the PDF domain is broad and each tool maps to a distinct operation across reading, editing, conversion, security, and analysis. There is little redundancy, so the count is heavy but defensible rather than excessive.
The surface covers a full PDF lifecycle: creation, reading, OCR, editing, page organization, forms, redaction, protection, conversion, extraction, comparison, and tamper detection. Notable gaps include annotations/comments, digital signature verification/signing, metadata editing, and image extraction, but these are minor relative to the core workflows.