Log inSubmit server
MinerU Open MCP logo

MinerU Open MCP

Parses PDFs, Office files, images, and HTML into Markdown through the MinerU API over MCP.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install MinerU Open MCP

Pick your client, then add it in one click or paste the config.

STDIOPython
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "mineru-open-mcp": {
      "command": "uvx",
      "args": [
        "mineru-open-mcp"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Automated check passed— started and listed 2 tools correctly (25d ago).

Overview

MinerU Open MCP MCP server lets MCP-compatible clients send documents to the MinerU API for parsing. It accepts PDFs, images, Word, PowerPoint, Excel, and HTML, returning Markdown for agent workflows. Flash mode does not require an API key and is limited to 10 MB and 20 pages, making it suitable for quick document extraction. Use the server when an agent needs to turn unstructured files or web content into text it can process.

Use cases

  • Parse research PDFs for an AI agent
  • Extract tables and formulas from Office documents
  • Convert scanned images into Markdown
  • Feed multilingual documents into a RAG workflow

Key features

  • Parse PDFs, Office files, images, and HTML
  • Return document content as Markdown
  • Support OCR and VLM-based extraction
  • Handle tables, formulas, and complex layouts
  • Process up to 20 pages without a token in Flash mode

Tools2

What MinerU Open MCP lets your agent do. Expand a tool for its parameters.

Verified live~819 tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 1d ago

Recent health check succeeded.

8/8 checks healthy over the last 40d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 13d ago via OSV.dev · mineru-open-mcp (PyPI)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars229Stargazers on the source repository.
Last commit1mo agoMost recent push to the default branch.
Availability100%Our rolling endpoint + install checks that succeeded.
Install checkPassedOur sandbox started it and listed its tools.
Tools exposed2Callable tools this server registers over MCP.
Directory activity3 viewsConfig copies, upvotes, and views on AllMCPs.

How MinerU Open MCP works

What MinerU Open MCP does

MinerU Open MCP MCP server connects MCP-compatible AI clients to the MinerU Open API for document parsing. It is part of the MinerU-Ecosystem repository, which also contains command-line tools, SDKs, and integrations for other frameworks. The MCP path is intended for agents that need document content as a native tool rather than through a separate application.

The server can process PDFs, images, DOC and DOCX files, PPT and PPTX presentations, XLS and XLSX spreadsheets, and HTML. Parsed content is returned as Markdown. The underlying MinerU service can also produce JSON, but the listing specifically describes the MCP use case as document parsing into Markdown.

How it works

An MCP-compatible client invokes the server, which uses MinerU’s API to analyze the supplied document. MinerU supports a VLM and OCR-based processing approach, with recognition for 109 languages. Its parsing capabilities include formulas represented as LaTeX, tables represented as HTML, scanned documents, handwriting, multi-column layouts, and tables that continue across pages. The output is arranged in reading order, with automatic header and footer removal described among MinerU’s core capabilities.

The repository identifies the MCP server as a Python component and lists Claude Desktop, Cursor, and Windsurf as supported client paths. The wider ecosystem also provides a web-crawling operation through its CLI and SDKs, but the supplied MCP details do not specify whether that operation is exposed by this server.

Setup and configuration

It does not include the MCP installation command, configuration schema, launch command, or required environment-variable names, so those details should be checked in the repository before deployment.

Flash or Quick Parse mode is described as requiring no API token. This mode is rate-limited by IP and uses a fixed lightweight pipeline model. The alternative Precision Extract API requires a token, supports selectable models such as pipeline, vlm, and MinerU-HTML, and is intended for larger or higher-fidelity jobs.

Tools and capabilities

MinerU Open MCP MCP server is suited to:

  • Parsing PDFs and scanned documents into Markdown
  • Extracting content from Office files, images, and HTML
  • Preserving formulas, tables, and complex document layouts
  • Handling multilingual documents with MinerU’s OCR support
  • Supplying parsed content to agent workflows through MCP

The Precision Extract API supports files up to 200 MB and 200 pages, with batches of up to 200 files. Quick Parse or Flash mode handles one file at a time, with a 10 MB and 20-page limit.

Limitations and notes

Quick Parse does not require authentication, but it is IP rate-limited and has stricter size, page, and single-file limits. Precision extraction requires an API token. The available API modes differ in model selection and output formats: Quick Parse returns Markdown, while Precision Extract can return Markdown, JSON, or ZIP and can optionally export DOCX, HTML, or LaTeX.

Confirm those implementation details in the repository before integrating MinerU Open MCP MCP server into an automated pipeline.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about MinerU Open MCP

MinerU Open MCP is an MCP server that connects AI clients to the MinerU Open API for document parsing. Its main tools let agents submit PDFs, images, Office files, or HTML and receive parsed Markdown. Flash mode works without an API key, while Precision Extract requires a token and supports larger jobs.

No, Flash or Quick Parse mode does not require an API key. It is IP rate-limited and limited to one file of up to 10 MB and 20 pages. Precision Extract does require an API token and provides higher limits, batch processing, and additional model and output-format options.

Claude Desktop, Cursor, and Windsurf are explicitly named as supported clients.

MinerU Open MCP can parse PDFs, images, DOC and DOCX files, PPT and PPTX files, XLS and XLSX files, and HTML. The documented MCP use case returns Markdown, while the underlying Precision Extract API also supports JSON, ZIP, and optional DOCX, HTML, or LaTeX exports.

Related MCP servers

All in Search & Data Extraction All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewMinerU Open MCP AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/mineru-open-mcp?style=directory)](https://allmcps.com/mcp/mineru-open-mcp)
HTML Embed
<a href="https://allmcps.com/mcp/mineru-open-mcp"><img src="https://allmcps.com/api/badge/mineru-open-mcp?style=directory" alt="MinerU Open MCP on AllMCPs" /></a>