webstractor

Agent ready

The web, distilled intoagent-ready context.

Search and extract the public web as clean Markdown or structured JSON. One cache-first GET API, plus a hosted MCP server built for agents.

Try the live API Read the quickstart

[ No account · 10 free requests daily · Cache hits are free ]

Loading example…
Open raw ↗
Loading the cached example…
Output format
Search across sources

Search the public web.

Find webpages, current news, openly licensed images, public videos, and places through one consistent interface.

View docs
Public URL extraction

Turn any public URL into context.

Submit the normal page a person would open. Receive readable Markdown or normalized schema-v1 JSON.

View docs
Loading example…
Open raw ↗
Loading the cached example…
Output format
Loading example…
Open raw ↗
Loading the cached example…
Output format
Current market data

Give agents current market context.

Look up a public market symbol and receive its latest snapshot with a bounded price history.

View docs
GET-only public API30-day extraction cacheSchema v1 normalized output

Purpose-built adapters

One clean interface for the public web.

Works with these supported platforms

Public API

Clean web data in one request.

Search the web, news, images, videos, places, and market data, then use /api/extract for clean page and product content.

curl --get \
  'https://webstractor.com/api/extract' \
  --data-urlencode 'url=https://www.amazon.com/dp/B09B8V1LZ3' \
  --data-urlencode 'format=json'

Hosted MCP server

Give agents search and extraction.

Connect one MCP server. Agents can discover webpages, news, images, videos, places, stocks, and market movers, then read clean content with extract_public_url.

https://webstractor.com/mcp
Read the MCP docs →

Simple pricing

Pay only for what you extract.

Prepaid credits are the hard usage cap. Add funds when you need them, without a required monthly plan.

Pay as you go

$0.49

/ 1,000 credits

Each extraction, web search, news search, image search, video search, place search, or finance request counts as one credit.

Included

  • Markdown or normalized JSON
  • Every supported source and endpoint
  • Cache hits are not billed
  • Credits never expire
  • 60 API calls per minute
Claim 1,000 free credits

Need the full details? View pricing and FAQs.