Crawl

Turn any website into clean, LLM-ready markdown for your agents and pipelines. Hanzo Crawl scrapes single pages or crawls whole sites, and its API is Firecrawl-compatible — point an existing Firecrawl client at api.hanzo.ai and it just works.

Scrape a Page

Fetch one URL and get back clean markdown (scripts, nav, and boilerplate stripped):

curl -X POST https://api.hanzo.ai/v1/scrape \
  -H "Authorization: Bearer hk-..." \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com",
    "formats": ["markdown", "html"]
  }'

{
  "success": true,
  "data": {
    "markdown": "# Example Domain\n\nThis domain is for use in ...",
    "metadata": { "title": "Example Domain", "statusCode": 200 }
  }
}

Crawl a Site

Start an asynchronous crawl over a site, then poll the job for results:

# Start the crawl → returns a job id
curl -X POST https://api.hanzo.ai/v1/crawl \
  -H "Authorization: Bearer hk-..." \
  -H "Content-Type: application/json" \
  -d '{ "url": "https://docs.example.com", "limit": 100 }'

# Poll status + collected pages
curl https://api.hanzo.ai/v1/crawl/JOB_ID \
  -H "Authorization: Bearer hk-..."

Use POST /v1/map for a fast URL list of a site without fetching page bodies.

Feed Your Agents

Crawl output is clean markdown, ready to embed or hand to a model — pair it with the LLM Gateway for retrieval, or expose it to agents as an MCP tool. For structured text from uploaded files (PDF, DOCX, XLSX), use the extract service instead.

Related

• MCP — give agents web access as a tool
• AI Studio · Marketplace
• API Keys · API Reference