# ComPDF MCP server

> Lets an AI client convert, edit, and parse PDFs and images, extracting structured text, tables, and data.

- Repository: https://github.com/ComPDFKit/compdf-mcp
- Page: https://hostedmcp.org/servers/compdfkit/compdf-mcp
- Official: yes
- Transport: —
- Stars: 98
- License: Apache-2.0

## Environment variables

| Name | Required | Secret | Description |
|---|---|---|---|
| `COMPDF_API_BASE_URL` | no | no |  |
| `COMPDF_API_TIMEOUT_SECONDS` | no | no |  |
| `MCP_ALLOWED_HOSTS` | no | no |  |
| `MCP_HOST` | no | no |  |
| `MCP_PORT` | no | no |  |
| `MCP_RATE_LIMIT_PER_MINUTE` | no | no |  |

## FAQ

### What can ComPDF MCP do with PDF files?

It converts PDFs to formats like Word, Excel, Markdown, and JSON, performs editing tasks such as merge, split, compress, and encrypt, and parses or extracts structured data from PDFs and images.

### Can ComPDF MCP extract data from scanned PDFs and images?

Yes, it runs OCR on scanned PDFs to produce searchable PDFs and extracts text, tables, and fields from PDFs and images into structured output.

### What tools are in the AI module endpoint?

The AI module exposes two tools, document_parse and document_extract, for intelligent parsing and data extraction.