MCP Tools

QuickCrawl MCP server provides 5 tools for web scraping and search.

Available Tools

Tool	Description
scrape	Scrape a single URL
crawl	Start an async BFS crawl
check_crawl_status	Check crawl job status by ID
site_map	Discover URLs without scraping (sitemap-aware)
search	Search SearXNG and optionally scrape results

Tool: scrape

Scrape a single URL with configurable formats and rendering.

{
  "url": "https://www.notion.so/",
  "formats": ["markdown"],
  "renderMode": "browser"
}

Parameters:

• url (string, required) — URL to scrape
• formats (string[]) — Output formats: markdown, html, links, json
• renderMode (string) — "auto", "browser", "http"
• waitFor (int) — Milliseconds to wait after JS rendering
• includeTags (string[]) — CSS selectors to include
• excludeTags (string[]) — CSS selectors to exclude
• cssSelector (string) — Extract content from specific CSS selector

Tool: crawl

Start an async BFS crawl of a website.

{
  "url": "https://example.com",
  "maxDepth": 2,
  "maxPages": 50,
  "formats": ["markdown"]
}

Parameters:

• url (string, required) — Starting URL
• maxDepth (int) — Maximum crawl depth (0-100). Default: 2
• maxPages (int) — Maximum pages (1-100). Default: 100
• formats (string[]) — Output formats for each page: markdown, html, links
• renderMode (string) — Render mode override
• waitFor (int) — Wait time after each page load

Tool: check_crawl_status

Check the status of a crawl job by its ID.

{
  "id": "crawl-1748899200000000000"
}

Parameters:

• id (string, required) — Crawl job ID

Tool: site_map

Discover all URLs on a website without scraping content. Uses sitemap.xml by default.

{
  "url": "https://example.com",
  "maxDepth": 2
}

Parameters:

• url (string, required) — Starting URL
• maxDepth (int) — Maximum depth (0-100). Default: 2
• useSitemap (bool) — Use sitemap.xml as seed URLs. Default: true
• timeout (int) — Timeout in ms. Default: 30000

Tool: search

Search SearXNG and optionally scrape result URLs.

{
  "query": "golang web scraping",
  "scrape": true,
  "formats": ["markdown"]
}

Parameters:

• query (string, required) — Search query
• scrape (bool) — Scrape each result URL. Default: false
• formats (string[]) — Output formats when scraping: markdown, html, links, json
• timeRange (string) — day, week, month, year
• region (string) — Language/region code
• page (int) — Page number (1-based)
• useBM25 (bool) — Use BM25 scoring. Default: false
• renderMode (string) — Render mode for scraping results