HTML Extract

Extract content from HTML pages and files using MinerU. Converts HTML to clean, structured Markdown preserving headings, lists, tables, and text hierarchy. F...

Security Scan

The skill's declared binary requirement (mineru-open-api) and single MINERU_TOKEN credential match its HTML extraction purpose and the SKILL.md instructions; nothing appears disproportionate or unrelated.

Purpose & Capability

The name/description (HTML extraction via MinerU) align with the declared runtime requirement (mineru-open-api) and the single required env var (MINERU_TOKEN). Requiring a MinerU CLI and token is expected for this functionality.

Instruction Scope

SKILL.md contains explicit commands using mineru-open-api (extract, crawl) and only references local HTML files, URLs, and the MINERU_TOKEN. It does not instruct reading unrelated system files, other environment variables, or exfiltrating data to unexpected endpoints.

Install Mechanism

Installers are npm (mineru-open-api) and go install from the GitHub repo — these are standard package sources. Installing third-party packages runs remote code at install/runtime, so verify the npm package and GitHub repository are the legitimate MinerU project before installing.

Credentials

Only one credential (MINERU_TOKEN) is required and is declared as primaryEnv. This is proportionate to a CLI that calls a remote MinerU API. No unrelated secrets or broad filesystem config paths are requested.

Persistence & Privilege

The skill does not request always:true or other elevated persistence. It is user-invocable and allows normal autonomous invocation, which is the platform default and reasonable for this capability.

Assessment

This skill is internally consistent with its stated purpose, but you should verify the mineru-open-api package before installing: check the npm package page and the GitHub repo linked from the MinerU homepage (https://mineru.net / https://github.com/opendatalab). Treat MINERU_TOKEN as a secret (do not reuse highly privileged credentials), create a token with least privilege if possible, and rotate it if you later stop using the skill. If you're cautious, install the CLI in an isolated environment (container or VM) and inspect its behavior (requests it makes) before using with sensitive data.

Extract text and content from local HTML files to Markdown using MinerU. For live web page URLs, use mineru-open-api 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

# Extract from a local HTML file (requires token)
mineru-open-api extract page.html -o ./out/

# Extract from a remote HTML URL (requires token)
mineru-open-api extract https://example.com/page.html -o ./out/

# Extract web page content via crawl (requires token)
mineru-open-api crawl https://example.com/article -o ./out/

# With language hint
mineru-open-api extract page.html --language en -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 remote HTML URL
• HTML requires extract (token required) — not supported by flash-extract
• For live web pages, use mineru-open-api crawl <URL> (also requires token)
• Language hint with --language (default: ch, use en for English)

Notes

• HTML is NOT supported by flash-extract — always use extract or 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