tomevault-io Web Fetcher
Extract web page content as clean text/markdown from a given URL.
Usage
Run the fetch script:
python3 <skill-path>/scripts/fetch.py <url>
Save to file:
python3 <skill-path>/scripts/fetch.py <url> -o output.md
Fallback Chain
The script tries these sources in order, falling back on failure:
- Jina Reader (
https://r.jina.ai/{url}) — best markdown quality - defuddle.md (
https://defuddle.md/{url}) - markdown.new (
https://markdown.new/{url}) - Raw HTML — direct fetch as last resort
Progress and errors are printed to stderr. Only the final content goes to stdout.
Source: kenmick/skills — distributed by TomeVault.