Everything you need to read the web

Scrape

Fetch any URL and get clean markdown, HTML, links, screenshots, or structured JSON. Handles headless rendering, PDF parsing, and page actions.

Watch mode

Subscribe to a URL and receive a webhook when its content meaningfully changes. Backed by an adaptive polling engine and a semantic differ rather than a hash check.

Map

Discover every accessible URL on a site in a single API call. Useful for planning a crawl or building a complete site index.

Crawl

Recursively scrape an entire site or subtree. Set depth, page limits, and URL patterns. Results stream in as the worker progresses.

Batch scrape

Submit up to 1,000 URLs in a single job. Each URL is scraped independently and results are available via polling.

Extract

Pass a URL and a JSON schema. Bluefield fetches the page and uses an LLM to pull out structured data that matches your schema exactly.

Search

Run a web search and optionally scrape the result pages in one call. Returns ranked pages with clean content ready for LLMs or agents.

Attestation

Every live fetch returns a signed provenance manifest. Verify authenticity and timestamp independently, without trusting Bluefield. Cached responses and PDFs are unsigned in v1.