pdfluent-mcp
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PDFLUENT_MCP_ROOTS | No | Allowed root directories the server may read and write inside, separated by ':' (';' on Windows). Repeatable equivalent of the --root option. A server with no root from any source (--root, this variable, or MCP roots offered by the client) answers every call with a message saying so. | |
| PDFLUENT_MCP_MAX_BYTES | No | The largest document the server opens. Equivalent of the --max-bytes option. | 100 MiB |
| PDFLUENT_MCP_MAX_PAGES | No | How many pages one call returns, so a long document is read in windows through the pages argument. Equivalent of the --max-pages option. | 200 |
| PDFLUENT_MCP_MAX_OUTPUT_CHARS | No | The longest result one call returns, after which it is cut and says so. Equivalent of the --max-output-chars option. | 200000 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
Read and fill local PDF files inside the allowed roots. Pass absolute paths. Use extract_markdown to read or quote a document, extract_text for the plain text layer, extract_tables for table data as rows, extract_blocks when you need the page and box of what you cite, get_metadata before reading a document you do not know, search to find the pages that mention something, and list_form_fields then fill_form to complete a form into a new file. Page numbers are 0-based in every argument and result. A page listed as needing vision has no text layer: nothing the text tools return for it is its content.
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 |
|---|---|
| extract_markdownA | A local PDF as markdown, in reading order. Headings become Args:
path: The PDF's absolute path on this machine.
pages: The 0-based pages to return; omit for the whole document, up to
the server's page limit. The whole document is still read, because
what a block is gets decided against the rest of it.
quality: |
| extract_textA | The plain text layer of a local PDF, page by page. Each page starts with Args: path: The PDF's absolute path on this machine. pages: The 0-based pages to return; omit for the whole document, up to the server's page limit. |
| extract_tablesA | The tables of a local PDF as rows of cell text.
Args: path: The PDF's absolute path on this machine. pages: The 0-based pages to look at; omit for the whole document, up to the server's page limit. |
| extract_blocksA | A local PDF as blocks, in reading order, with where each one came from. Each block has
Args: path: The PDF's absolute path on this machine. pages: The 0-based pages to return; omit for the whole document, up to the server's page limit. include_furniture: Keep running heads and page numbers (kinds header and footer); they are left out by default because they repeat on every page and crowd out the content. |
| get_metadataA | What a local PDF is, before reading it.
Args: path: The PDF's absolute path on this machine. |
| searchA | The pages of a local PDF that mention
Args: path: The PDF's absolute path on this machine. query: The text to look for, as written in the document. |
| list_form_fieldsA | The interactive form fields of a local PDF.
Args: path: The PDF's absolute path on this machine. |
| fill_formA | Fill the form fields of a local PDF and write the result to a new file.
Args: path: The absolute path of the PDF to fill. values: Field name to new value. out_path: The absolute path to write the filled PDF to. overwrite: Replace a file that already exists at out_path. |
| vision_adviceA | Which pages of a local PDF are worth sending to a vision model, and why.
Args: path: The PDF's absolute path on this machine. |
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 9 tools
Each tool targets a clearly distinct output or operation: markdown, plain text, tables, blocks, metadata, search, vision advice, and form listing/filling. Although several tools extract from the same PDF, their descriptions make the boundaries explicit and an agent can reliably choose between them.
All names use snake_case consistently, which is a clear convention. However, verbs are not fully uniform: extract_* dominates, but vision_advice is noun-oriented and search is a bare verb, so the set is mostly rather than perfectly consistent.
Nine tools is well-scoped for a PDF extraction and form-filling server. Each tool has a clear role and none appears redundant or merely decorative.
The surface covers the main PDF reading lifecycle: metadata, search, multiple extraction formats, vision triage, and form inspection/filling. Minor gaps remain, such as no direct page rendering or image extraction tool, but core workflows are supported.