Site Fetcher

Site Fetcher (sitefetch)

Fetch an entire website (or specific pages) as clean readable text — ready for AI analysis.

Quick Usage

# Fetch entire site
sitefetch https://example.com -o site.txt

# Fetch only blog posts
sitefetch https://example.com --match '/blog/**' -o blog.txt

# Fetch with higher concurrency (faster)
sitefetch https://example.com --concurrency 10 -o site.txt

# Fetch specific section
sitefetch https://docs.example.com --match '/api/**' -o api-docs.txt

When to Use

• Competitor analysis: fetch their site, analyze content strategy
• Content audit: get all pages as text, find gaps
• SEO research: extract all content from a domain
• Documentation gathering: pull docs site into one file
• Before/after comparison: fetch site, make changes, fetch again

Output

Creates a single text file with all pages concatenated. Each page is separated with its URL. Content is extracted using Mozilla Readability (same as Firefox Reader View) — no ads, navs, footers.

Options

• -o, --output <file> — output file path (default: stdout)
• --match <glob> — only fetch pages matching glob pattern
• --concurrency <n> — parallel requests (default: 5)

Examples

Analyze competitor blog

sitefetch https://competitor.com --match '/blog/**' -o /tmp/competitor-blog.txt
# Then read and analyze the file

Full site content audit

sitefetch https://mysite.com -o /tmp/mysite-content.txt
# Analyze for thin content, duplicates, missing topics

Grab documentation

sitefetch https://docs.library.com -o /tmp/docs.txt
# Use as reference for implementation

Notes

• HTTP only (no JavaScript rendering) — for JS-heavy SPAs use agent-browser instead
• Respects robots.txt
• Works best on content-heavy sites (blogs, docs, marketing pages)
• For single interactive pages, prefer agent-browser