grass-html

Extract Markdown and structured metadata from HTML.

Install

pip install grass-html

Usage

from grass_html import extract, to_markdown, to_json

result = extract(html_string)
print(result.markdown)
print(result.degraded)   # True when extraction coverage is low
print(result.coverage)   # ratio vs full-page markdown

md = to_markdown(html_string)
doc = to_json(html_string)

Features

• Single extraction profile: Markdown body + metadata
• Links and images always preserved
• Degradation detection via coverage threshold
• Structural verification utilities (structural_report, boilerplate_probes)

CLI

grass-html path/to/file.html

Status

Early stage. API may change between minor releases.