Web scraping API for AI agents

Web scraping API for clean, AI-ready Markdown.

Give your AI agents readable web content, not raw HTML. Search for sources, extract a known URL with Fetch or Render, and capture screenshots when your workflow needs a visual record.

Try a URL — no signup Get an API key

Core Feature

Use a browser only when the page needs one.

For a known URL, use Fetch when the useful content is already in HTML. Choose Render when JavaScript loads the content or browser behavior matters. Both paths return clean Markdown in a consistent crawl response.

Illustrative, shortened response based on our URL to Markdown · Fetch page. It is not a live crawl.

page crawlillustrative output

API response exampleSample JSON

{
  "ok": true,
  "requested_url": "https://anycrawler.com/crawler/page/fetch/",
  "canonical_url": "https://anycrawler.com/crawler/page/fetch/",
  "final_url": "https://anycrawler.com/crawler/page/fetch/",
  "status_code": 200,
  "credits_used": 2,
  "results": {
    "title": "URL to Markdown API for HTML Pages",
    "markdown": "# Convert a URL into clean Markdown—without a browser.\n\nUse Fetch for pages whose content is available in the HTML response."
  }
}

Compare page crawl modes URL to Markdown · Fetch JavaScript Web Scraping API

Core Feature

Add a web search API before crawling

Start from a query, discover candidate pages, news, images, videos, or scholarly sources, then send only the strongest results into fetch, render, or screenshot workflows.

pagenewsmore

What is AnyCrawler?page

AnyCrawler is web access infrastructure for AI systems that need reliable live context.

AnyCrawler Product Overviewpage

See search, crawl, render, screenshots, credits, and workspace controls in one place.

AnyCrawler API Documentationpage

Read request formats, response fields, and examples for building on the API.

AnyCrawler Consolepage

Choose a search channel Web Search API Image Search API News Search API Video Search API Scholar Search API

Context Control

Send the page content your model needs—not the surrounding HTML.

AnyCrawler returns readable Markdown for the next step in your AI workflow. Use theFetch API when the page content is in HTML, or the Render API when JavaScript must load it.

HTMLPage + UI

MarkdownMain text

Agent Skill

For AI agents

Plug live web access into agents.

Give agents a managed web access layer for search, crawl, render, screenshots, and structured outputs without building custom browser infrastructure.

Setup stays short: install the skill, follow the documented flow, and give the agent a reliable way to read the live web when a workflow needs current context.

Skill-first setup for agent workflows
Search, crawl, render, and screenshot capabilities
Structured outputs ready for the next tool step

Install Skill

Developer

For builders

Build on a production web data API.

Use stable endpoints, clear response shapes, API keys, usage logs, and credit-based billing to move from prototype to production without maintaining crawler infrastructure.

Skill docs and API docs stay aligned, so both agent callers and engineers can share the same underlying web access layer.

Concise API docs with copy-paste examples
Consistent auth, credit rules, and response shape
Production path for search, fetch, render, and screenshots

Read API Docs

From the Blog

Put web access into practice.

The latest guides to finding sources, extracting content, and building web-aware AI agents.

View all articles

Explore practical web access guides. Browse the Blog →

Pricing

Start with 5,000 free credits

Choose the operation your page needs. Base credit costs differ for search, fetch, render, and screenshots.

Base credits by operation

Page search2 credits per returned result
Render request10 credits base
Fetch request2 credits base
Viewport screenshot20 credits
Full-page screenshot50 credits

What 5,000 credits can cover

Each example uses the free plan's one-time credits for one operation at its base cost:

Fetch only2,500 requests
Render only500 requests
Search only, 10 results each250 searches
Viewport screenshots only250 screenshots
Full-page screenshots only100 screenshots

Optional fields and cache behavior can change actual credit use.

See plans and full pricing Try a URL — no signup

FAQ

Infrastructure questions for technical buyers

Short answers for teams evaluating AnyCrawler as the web access layer behind AI agents, RAG systems, research tools, and automation products.

What is AnyCrawler?

AnyCrawler is a web scraping API for AI agents and data products. One production API covers web discovery, HTML fetches, browser rendering, Markdown extraction, screenshots, API keys, request logs, and credit-based usage controls.

What can the AnyCrawler web scraping API do?

It can search for source URLs, fetch HTML-first pages, render JavaScript pages in a browser, return Markdown and selected metadata, and capture page screenshots. Each task has its own endpoint or explicit mode so an integration does not pay for browser work when a simple fetch is enough.

How does flexible billing work?

AnyCrawler uses credits so different web access tasks follow rules that match the work performed. Search, fetch, render, and screenshots each have explicit credit costs, allowing teams to combine capabilities without forcing every request into one flat model.

When should I use fetch, render, search, or screenshots?

Use search when the target source is unknown, fetch when useful content is available in HTML, render when JavaScript or browser behavior matters, and screenshots when a workflow needs visual evidence alongside structured output.

Who is AnyCrawler built for?

AnyCrawler is built for teams shipping AI agents, RAG systems, research tools, monitoring products, and automation workflows that need current web data without maintaining separate search, crawler, browser, parser, and usage systems.