convert_pdf_urlLABSCO SUMMARY
A PDF becomes text an agent can actually work with.
The conversion keeps the document's structure instead of flattening it, and the images come out as separate files — which is the difference between a Markdown file you can edit and a wall of extracted characters.
What it is
A PDF-to-Markdown converter built on Mistral AI's OCR. It takes a local file or a remote URL, and writes the converted Markdown and the extracted images into a directory you choose.
What you get
- Local PDF files converted to Markdown, in batches rather than one at a time
- Remote PDFs converted from their URL, without downloading them first
- Document structure preserved — headings, paragraphs and lists come through as Markdown rather than as flat text
- Images extracted from the PDF and saved alongside the output
- The output directory set on the command line, so converted files land where the rest of your work is
Requirements
Python 3.10 or newer with uv, and a Mistral AI API key — supplied in a .env file or in the client config. Documents are sent to Mistral for recognition, and the converted files are written to the output directory, which defaults to ./downloads.
Setup effort
Build from source — git clone https://github.com/zicez/mcp-pdf2md.git
USE CASES
01
Getting a report into a form you can edit
Convert the PDF and get Markdown with its headings and lists intact, ready to paste into a document or a repository.
02
Reading a paper you only have a link to
Pass the URL and let the server fetch and convert it, skipping the download-then-open step entirely.
03
Turning a pile of scanned documents into searchable text
Point it at several local files in one pass — batch conversion is what makes an archive of scans searchable rather than a folder you avoid.
04
Keeping the figures out of a document
Images are extracted and saved as files next to the Markdown, so a chart from a report can be reused without a screenshot.
2 tools
Listed in project docs. Your client may show a different list.
convert_pdf_fileBEFORE YOU INSTALL
What to know first
Tools you get
2 tools named in the project's documentation.
Requires
- Python 3.10+
- MISTRAL_API_KEY in your environment
Where it runs
Both documented setups run the server on your machine.
INSTALLATION
Connect your client
You clone this one yourself.
Server setup
TERMINAL
git clone https://github.com/zicez/mcp-pdf2md.git
cd mcp-pdf2md
uv syncMCP Clients
Claude Desktop — Windows, macOS/Linux
JSON
{
"mcpServers": {
"pdf2md": {
"command": "uv",
"args": [
"--directory",
"<YOUR_PATH_TO_MCP_PDF2MD>",
"run",
"pdf2md",
"--output-dir",
"<YOUR_PATH_TO_OUTPUT>"
],
"env": {
"MISTRAL_API_KEY": "<YOUR_MISTRAL_API_KEY>"
}
}
}
}
