ocrmypdf-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@ocrmypdf-mcp OCR the scanned PDF at ~/Downloads/scan.pdf"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
ocrmypdf-mcp
A minimal MCP server that exposes ocrmypdf as a single tool, ocr_pdf, so Claude can OCR scanned PDFs and then hand them to markitdown (or any text tool) for downstream work.
Why this exists: the obvious "just call ocrmypdf" approach falls over on Windows with the Microsoft Store (MSIX) build of Claude Desktop, because MSIX launches MCP servers with a stripped-down PATH that doesn't include Tesseract or Ghostscript. This server auto-detects the standard Windows install locations and prepends them to PATH at startup, so OCR Just Works without futzing with system environment variables.
Works on Linux and macOS too — the PATH augmentation is a no-op outside Windows.
Prerequisites (Windows)
Two system installers, then pip install.
1. Tesseract OCR
UB-Mannheim build (the standard Windows distribution): https://github.com/UB-Mannheim/tesseract/wiki
Accept the default install location (C:\Program Files\Tesseract-OCR). Add language packs during install if you need anything beyond English.
2. Ghostscript
AGPL release for Windows (free): https://www.ghostscript.com/releases/gsdnld.html
Accept the default install location (C:\Program Files\gs\gs<version>\).
3. Verify (optional)
tesseract --version
gswin64c --versionIf either says "not recognized," reopen PowerShell so it picks up the updated PATH, then retry.
Related MCP server: pdf-filler-recursive-simple-mcp
Install the server
git clone https://github.com/jcm4TX/ocrmypdf-mcp
cd ocrmypdf-mcp
pip install --user .This installs ocrmypdf, the mcp SDK, and the ocrmypdf-mcp executable. On Windows it lands at:
C:\Users\<you>\AppData\Roaming\Python\Python313\Scripts\ocrmypdf-mcp.exeWire it up in Claude Desktop
Edit claude_desktop_config.json. On the MSIX (Microsoft Store) build of Claude Desktop, the path is:
%LOCALAPPDATA%\Packages\Claude_pzs8sxrjxfjjc\LocalCache\Roaming\Claude\claude_desktop_config.jsonOn the regular non-MSIX installer it's:
%APPDATA%\Claude\claude_desktop_config.jsonAdd an ocrmypdf-mcp entry under mcpServers:
{
"mcpServers": {
"ocrmypdf-mcp": {
"command": "C:\\Users\\<you>\\AppData\\Roaming\\Python\\Python313\\Scripts\\ocrmypdf-mcp.exe",
"args": []
}
}
}Then fully quit Claude Desktop — right-click the tray icon and pick Quit, not just close the window — and relaunch.
Verify it loaded
In a new chat, ask "what MCP tools do you have for OCR?" — Claude should report ocr_pdf. If not, check the server log:
%LOCALAPPDATA%\Packages\Claude_pzs8sxrjxfjjc\LocalCache\Roaming\Claude\logs\mcp-server-ocrmypdf-mcp.logTool API
ocr_pdf(input_path, output_path?, language?, force_ocr?, deskew?)
Arg | Type | Default | Meaning |
| str | required | Absolute path to input PDF |
| str |
| Where to write the OCR'd PDF |
| str |
| Tesseract language code; join multiple with |
| bool |
| Re-OCR pages that already have a text layer |
| bool |
| Straighten skewed pages before OCR |
Default behavior: pages without an existing text layer get OCR'd, pages that already have text pass through unchanged. Safe to run on mixed PDFs.
Typical workflow
You hand Claude a scanned PDF path.
Claude calls
ocr_pdf(input_path="...").Claude calls
markitdown.convert_to_markdownon the resulting-ocr.pdf.Claude reads the markdown and answers your question.
Known limitations
The MCP protocol enforces a per-request timeout (~4 minutes in current Claude Desktop). Large multi-page documents may exceed this and surface as a client-side timeout even though the underlying
ocrmypdfprocess completes successfully — the output PDF will still be on disk. If you hit this regularly, split the input into smaller page ranges first.Complex multi-column scanned layouts (legal, probate, ledgers) can produce messy markdown when piped to markitdown afterward, because Tesseract interprets visual alignment as table structure. Post-processing the markdown to drop empty table-pipe rows recovers most of it.
License
Available Tools
1 toolocr_pdfA
Run OCR on a PDF to add a searchable text layer using Tesseract.
Use this when a PDF is scanned (no extractable text). After running, the output PDF has an invisible text layer and can be converted to markdown via markitdown, searched in Adobe / Windows Search, or copied from normally.
Args: input_path: Absolute path to the input PDF on the user's filesystem. output_path: Absolute path for the OCR'd output PDF. If omitted, defaults to "<input_stem>-ocr.pdf" in the same directory as input. language: Tesseract language code (e.g. "eng", "eng+spa", "deu"). Multiple languages are joined with "+". Default "eng". force_ocr: If True, OCR every page even if a text layer already exists. Use when the existing text layer is poor or junk OCR. Default False (pages with existing text are passed through unchanged). deskew: If True, straighten skewed pages before OCR. Default True.
Returns:
A dict with success (bool), output_path (str on success), and
error (str on failure).
Inputs: input_path (required), output_path, language, force_ocr, +1 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 1 tool
Only one tool exists, so there is no possibility of confusion or overlap with other tools.
With a single tool, naming is inherently consistent; 'ocr_pdf' clearly conveys the action and target.
A single tool for OCR seems thin; typically one might expect separate tools for checking text presence or converting formats, but the comprehensive parameter set partially justifies the minimal count.
The tool covers core OCR functionality with useful options (language, deskew, force_ocr), but lacks a companion tool to inspect existing text layers, a minor gap.
Related MCP Connectors
Hosted OCRmyPDF: add a searchable text layer to scanned PDFs and images, returned as PDF/A.
PDF tools for agents: compress to a size, Arabic + English OCR, Office conversion, JSON extraction.
PDF and image tools for AI agents: compress, merge, split, OCR, sign. Runs locally, never uploads.
PDF tools for AI agents: validate, extract, merge, split, compress, OCR, sign and redact.
Related MCP Servers
- FlicenseAqualityDmaintenanceExtracts text content from PDFs and images using Mistral's OCR API, enabling OCR capabilities in MCP-compatible clients like Cursor and Claude Desktop.18-
- AlicenseNot gradedqualityDmaintenanceProvides PDF analysis, data extraction, form filling, and comparison capabilities for Claude and Cursor, supporting both text and scanned documents with OCR.7 npm1MIT
- AlicenseNot gradedqualityDmaintenanceWraps Baidu's Unlimited-OCR model as MCP tools for Claude Code, enabling unlimited local OCR without API costs, supporting images and multi-page documents.MIT
- AlicenseAqualityCmaintenanceEnables OCR-enabled PDF to Markdown conversion in Claude Code, with per-page inspection and hybrid handling of text, scanned, mixed, and blank pages via a local vision-model endpoint.3MIT