skills.rest
blessonism
blessonismCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

mineru-extract

Convert URLs into Markdown and structured JSON via MinerU API.

55|5|Updated Feb 10, 2026
Search Tags:#ocr#pdf#markdown#html#api#mineru#extract
One-click install
npx skills add https://github.com/blessonism/openclaw-skills --skill mineru-extract
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mineru-extract
Source: https://github.com/blessonism/openclaw-skills/tree/main/mineru-extract
Command: npx skills add https://github.com/blessonism/openclaw-skills --skill mineru-extract

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill normalizes web content by converting HTML pages (for example WeChat articles) and direct document links (PDF/Office/image) into clean Markdown plus structured outputs for downstream processing.

Core Features & Use Cases

  • High-fidelity parsing: preserves layout, tables, formulas, and supports OCR for difficult sources where content is degraded or blocked.
  • URL-to-Markdown transformation: accepts single URLs or batched inputs and returns Markdown plus a structured JSON contract with output paths.
  • Use Case: Convert a batch of articles into Markdown for a knowledge base or content archive, enabling consistent downstream ingestion.

Quick Start

Submit a URL to MinerU and retrieve the resulting Markdown and metadata for downstream processing.

Frequently Asked Questions about mineru-extract

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I convert a web page URL to clean Markdown for a knowledge base?▼

To convert a web page URL to clean Markdown, this Skill calls MinerU's API to parse HTML pages like WeChat articles, downloads a ZIP archive, extracts the Markdown, and returns structured JSON output paths for downstream ingestion.

Can I extract tables and formulas from PDF documents using OCR?▼

Yes, you can extract tables and formulas from PDF documents using OCR. The process calls MinerU's API to support high-fidelity parsing, preserving layout, tables, and formulas even from degraded or blocked content sources.

What is the best way to batch convert articles into Markdown?▼

The best way to batch convert articles into Markdown is submitting URLs to this Skill, which processes batched inputs via MinerU, caches repeated requests to speed processing, and returns a JSON contract with output paths for consistent archiving.

Do I need a MinerU token to parse HTML pages and direct document links?▼

Yes, you need a MinerU token to parse HTML pages and direct document links. The Skill requires this token to authenticate API requests, process URLs, download the resulting ZIP archive, and extract the clean Markdown output.

What file formats are supported when converting web content to Markdown?▼

Supported file formats when converting web content include HTML pages, direct PDF links, Office documents, and images. The Skill calls MinerU's API to normalize these sources into clean Markdown and structured outputs.

Why does the PDF to Markdown conversion return a ZIP archive instead of text?▼

The PDF to Markdown conversion returns a ZIP archive because MinerU packages the extracted Markdown, structured outputs, and metadata together. The Skill automatically downloads, extracts this archive, and returns direct file paths within a JSON contract.

Related Skills

Explore More Content & Communication Skills→
opendatalabopendatalab
185

MinerU Document Extractor

Convert PDFs, images, and Office files into Markdown, HTML, or JSON.

Official
Intermediate
Lucius646Lucius646

karpathy-llm-wiki

Manage Markdown-based knowledge bases with LLM-driven ingestion, querying, and maintenance.

Community
Intermediate
while-basicwhile-basic

enact/firecrawl

Extract web page content as Markdown or structured data via the Firecrawl API.

Community
Advanced
Liangwei-zhangLiangwei-zhang

doc-reader

Extract text from PDFs and apply OCR to image-based pages.

Community
Intermediate