Web scraping and browser control, for agents that pay for themselves

Scrape any page or document to markdown, map and crawl whole sites, and drive a real browser. No API key. No signup. No monthly plan. Your agent answers a 402 and the call runs.

· pay per call, from $0.001 · USDC on Base (x402) · MCP server included · balance never expires

Try it without an account

curl -X POST https://api.oassis.dev/web/v1/scrape \
  -H 'content-type: application/json' \
  -d '{"url":"https://example.com","formats":["markdown","controls"]}'

That answers 402 with the exact price of that call. Pay it with a wallet and repeat the request, or send an API key and it is charged to a prepaid balance.

What it costs

Read a page	$0.001	per output asked for: markdown, html, links, controls…
Read a document	$0.002	PDF, Word, Excel or CSV by url, no browser
From the cache	$0.0002	a recent render, priced before you are charged
List a site's urls	$0.0003	from its sitemap
Open a browser session	$0.005	includes the first minute
Act on the page	$0.0005	click, type, select, scroll, wait
Search the web	$0.007	passed through at what the provider charges

Priced per call, quoted before any work happens, and refunded when the work does not. A request we cannot serve is never charged for.

What it does that a scraper usually does not

A map of the controls. Ask for controls and you get every link, button and field with its role, its readable name and where it sits on screen — the map an agent needs to decide what to click.
Browser sessions that stay open. Look, click, look again, without reloading and without losing state.
Documents by url. A PDF, Word, Excel or CSV comes back as markdown without opening a browser at all.
Every output in one call. markdown, html, links, controls, the accessibility tree, elements by selector, a screenshot, a PDF, or AI-extracted JSON.

For agents

MCP server: https://api.oassis.dev/mcp — initialize and tools/list are free, and a keyless client gets 10 reads a day, one browser session and 5 searches to try it.
OpenAPI: https://api.oassis.dev/openapi.json
Plain text for models: https://api.oassis.dev/llms.txt
Every route describes itself: GET https://api.oassis.dev/web/v1/scrape