0xMassi/webclaw

v0.2.0 — Major feature release

Document Extraction

webclaw now auto-detects and extracts content from document files:

• DOCX — Word documents parsed into markdown with headings preserved
• XLSX/XLS — Spreadsheets converted to markdown tables (multi-sheet support)
• CSV — Parsed with quoted field handling, output as markdown table

Auto-detected by Content-Type header or URL extension. Works in batch mode too:

webclaw https://example.com/report.docx
webclaw https://example.com/data.xlsx -f json
webclaw --urls-file mixed-urls.txt --output-dir ./results

HTML Output Format

webclaw https://example.com -f html

Returns sanitized HTML. Works with crawl, batch, and --output-dir (.html extension).

Multi-URL Watch

echo "https://site1.com/pricing
https://site2.com/status" > urls.txt
webclaw --urls-file urls.txt --watch --watch-interval 300 --webhook "https://discord.com/..."

Monitors all URLs in parallel. Reports aggregate changes per check.

Batch + LLM Extraction

webclaw --urls-file sites.txt --extract-prompt "get email and phone" --output-dir results

Combines batch fetching with LLM extraction. Processes URLs sequentially to respect rate limits.

Full changelog: https://github.com/0xMassi/webclaw/blob/main/CHANGELOG.md

Full Changelog: https://github.com/0xMassi/webclaw/compare/v0.1.7...v0.2.0