📖 Overview

MinerU-Ecosystem provides a full suite of tools, SDKs, and integrations built on top of the MinerU Open API. Whether you're building production pipelines, integrating with LangChain for RAG, or enabling AI agents to parse documents on the fly — this repository has you covered.

MinerU is an open-source, high-quality document extraction tool that converts unstructured documents (PDFs, images, Office files, etc.) into machine-readable Markdown and JSON.

🏗️ Repository Structure

MinerU-Ecosystem/
├── cli/                  # Command-line tool for document parsing
├── sdk/                  # Multi-language SDKs
│   ├── python/           #   Python SDK
│   ├── go/               #   Go SDK
│   └── typescript/       #   TypeScript SDK
├── langchain_mineru/     # LangChain document loader integration
├── mcp/                  # Model Context Protocol server (Python)
└── skills/               # AI agent skills (Claude Code, OpenClaw, etc.)

🔑 Supported APIs

All components in this repository support both API modes:

🚀 Quick Start

CLI (cli/)

A fast command-line tool for parsing documents directly from your terminal. Supports both Standard API and Quick Parse API.

Installation

Windows (PowerShell)

irm https://cdn-mineru.openxlab.org.cn/open-api-cli/install.ps1 | iex

Linux / macOS (Shell)

curl -fsSL https://cdn-mineru.openxlab.org.cn/open-api-cli/install.sh | sh

Usage

1. Flash Extract (no login, fast, Markdown only)

Great for quick previews. No Token needed. Limit: 10 MB / 20 pages per file.

mineru-open-api flash-extract report.pdf

2. Precision Extract (login required)

Supports large documents (200 MB / 600 pages), preserves layout and resources, multiple output formats.

# First-time setup: configure Token (or set MINERU_TOKEN env var)
mineru-open-api auth

# Extract and print Markdown to stdout
mineru-open-api extract paper.pdf

# Extract and save all resources (images/tables) to a directory
mineru-open-api extract report.pdf -o ./output/

# Export to other formats
mineru-open-api extract report.pdf -f docx,latex,html -o ./results/

3. Web Crawl

Convert web pages into high-quality Markdown.

mineru-open-api crawl https://www.example.com

4. Batch Processing

# Batch process all PDFs in the current directory
mineru-open-api extract *.pdf -o ./results/

# Batch process from a file list
mineru-open-api extract --list filelist.txt -o ./results/

Python SDK

Installation

pip install mineru-open-sdk

Usage

1. Flash Extract (no login, Markdown only)

Great for quick previews. No Token needed.

from mineru import MinerU

# Flash mode requires no Token
client = MinerU()
result = client.flash_extract("https://cdn-mineru.openxlab.org.cn/demo/example.pdf")

print(result.markdown)

2. Precision Extract (login required)

Supports large files, rich assets (images/tables), and multiple output formats.

from mineru import MinerU

# Get a free Token from https://mineru.net
client = MinerU("your-api-token")
result = client.extract("https://cdn-mineru.openxlab.org.cn/demo/example.pdf")

print(result.markdown)
print(result.images)  # Get the list of extracted images

Multi-language SDKs are also available: Go | TypeScript. See the sdk/ directory for details.

AI Agent Skills (skills/)

Pre-built skill for AI coding agents, enabling document extraction directly within agent workflows. The skill is wrapper by the mineru-open-api CLI and provides:

Skills Download

• OpenClaw — View skill details on ClawHub
• CDN Link — One-click download skill package
• Other AI agents like zeroclaw that also support skill/tool interfaces

MCP Server (mcp/)

A Model Context Protocol server implementation in Python, allowing MCP-compatible AI clients (such as Claude) to use MinerU's document parsing as a tool.

Configuration

Using uvx (recommended — always runs the latest version):

{
  "mcpServers": {
    "mineru": {
      "command": "uvx",
      "args": ["mineru-open-mcp"],
      "env": {
        "MINERU_API_TOKEN": "your_key_here"
      }
    }
  }
}

LangChain Integration (langchain_mineru/)

A LangChain Document Loader that turns PDFs and documents into LangChain-compatible Document objects with one line of code — ready to plug into RAG pipelines.

Installation

pip install langchain-mineru

Usage

from langchain_mineru import MinerULoader

loader = MinerULoader(source="demo.pdf")
docs = loader.load()

print(docs[0].page_content[:500])
print(docs[0].metadata)

Default is mode="flash" (no API token required). Switch to mode="precision" for higher fidelity with token auth.

Two parsing modes are available:

See the full documentation and RAG pipeline examples in langchain_mineru/.

📚 Documentation

📄 License

This project is licensed under the Apache License 2.0.