HTML Markdown

Convert HTML to Markdown using MinerU's document processing engine. Transforms HTML pages and files into clean, readable Markdown format. Features: HTML to M...

Security Scan

The skill is internally consistent: it wraps the mineru-open-api CLI to convert HTML to Markdown and only requests the MinerU token and the mineru-open-api binary that such a tool would reasonably need.

Purpose & Capability

Name/description match the requested artifacts: the skill requires the mineru-open-api binary and a MINERU_TOKEN, both of which are expected for a CLI that calls MinerU's service.

Instruction Scope

SKILL.md instructs the agent to run mineru-open-api extract/crawl and to set MINERU_TOKEN. It does not request other environment variables, read unrelated system files, or send data to unexpected endpoints beyond MinerU's service.

Install Mechanism

Installers are npm (mineru-open-api) and a go install from a GitHub repo — common and traceable mechanisms. As with any third-party package, installing a global npm package or go binary executes upstream code; review the npm package and GitHub repo before installing.

Credentials

Only MINERU_TOKEN is required and declared as the primary credential. That is proportionate for a hosted document-processing API. No unrelated credentials or config paths are requested.

Persistence & Privilege

always is false and the skill does not request system-wide configuration changes. It is an instruction-only skill that invokes an external CLI at runtime.

Assessment

This skill appears coherent: it calls the MinerU CLI and needs only MINERU_TOKEN. Before installing, verify the npm package and GitHub repo (https://github.com/opendatalab/MinerU-Ecosystem) to ensure you trust the upstream code. Treat MINERU_TOKEN like any secret: do not reuse high-privilege tokens, restrict its scope if possible, and run the CLI in a controlled environment when processing untrusted HTML. If you cannot or will not install third-party binaries, do not grant the skill access to an environment where mineru-open-api is available.

Convert HTML files to clean Markdown using MinerU. For local HTML files use extract; for live web pages use crawl.

Install

npm install -g mineru-open-api
# or via Go (macOS/Linux):
go install github.com/opendatalab/MinerU-Ecosystem/cli/mineru-open-api@latest

Quick Start

# Convert local HTML file to Markdown (requires token)
mineru-open-api extract page.html -o ./out/

# Convert remote HTML file to Markdown (requires token)
mineru-open-api extract https://example.com/doc.html -o ./out/

# Convert live web page to Markdown via crawl (requires token)
mineru-open-api crawl https://example.com/article -o ./out/

Authentication

Token required:

mineru-open-api auth             # Interactive token setup
export MINERU_TOKEN="your-token" # Or via environment variable

Create token at: https://mineru.net/apiManage/token

Capabilities

• Supported input: local .html file or HTTP/HTTPS URL
• Local .html file: use extract (token required)
• Live web page URL: use crawl (token required)
• HTML is NOT supported by flash-extract
• Language hint with --language (default: ch, use en for English)

Notes

• HTML input always requires token (no flash-extract support)
• For live web pages with JavaScript rendering, prefer crawl
• Output goes to stdout by default; use -o <dir> to save to a file or directory
• All progress/status messages go to stderr; document content goes to stdout
• MinerU is open-source by OpenDataLab (Shanghai AI Lab): https://github.com/opendatalab/MinerU