Only the content that matters
Main content mode drops navigation, headers and sidebars, so your agent reads the page and not the chrome.
Web scraping API · Built for agents
curl 'https://api.elisebot.dev/https://example.com/pricing'Try it without a key: single pages, up to 10 a minute.
claude mcp add --transport http elisebot https://api.elisebot.dev/mcpOr add https://api.elisebot.dev/mcp to any MCP client. It signs you in through the browser.
curl https://api.elisebot.dev/v2/batch/scrape \
-H "Authorization: Bearer $ELISEBOT_API_KEY" \
-H 'Content-Type: application/json' \
-d '{ "urls": ["https://example.com/a", "https://example.com/b"] }'Up to 50,000 URLs per request. Every result arrives in one file once the batch scrape ends.
Features
Main content mode drops navigation, headers and sidebars, so your agent reads the page and not the chrome.
Add the MCP server in one line. Your agent scrapes pages on its own and gets back Markdown it can read.
Pages behind bot checks and challenges come back like any other, with nothing to set up on your side.
Scrape at scale: follow the progress while the batch scrape runs, then download every result as one file. A webhook tells you when it is ready.
Get notified when a batch scrape starts and finishes. Every event is signed with your secret and retried until it lands.
Resend with the same Idempotency-Key and get the original batch scrape back instead of a duplicate, for zero credits.
Pay per page with prepaid credits. Failed pages are refunded.
curl https://api.elisebot.dev/v2/batch/scrape \
-H "Authorization: Bearer $ELISEBOT_API_KEY" \
-H "Idempotency-Key: nightly-2026-10-19" \
-H 'Content-Type: application/json' \
-d '{
"urls": ["https://example.com/a", "https://example.com/b"],
"responseOptions": { "format": "markdown", "onlyMainContent": true },
"webhook": { "url": "https://you.example/hooks/elisebot" }
}'