Search the web#

POST /v1/search

Find relevant sources, then optionally retrieve their page content in the same operation. Luni searches the public web directly. A dedicated provider key, when configured, is used instead.

Parameter	Type	Default	Description
`query`	string	Required	Search text, trimmed to 1–500 characters.
`limit`	integer	5	Maximum results, from 1 to 25.
`scrapeOptions`	object	Omitted	Add supported scrape settings to fetch content from the result URLs.

curl -X POST https://api.luni.gg/v1/search \
  -H "Authorization: Bearer $LUNI_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: search-sources" \
  -d '{"query":"public web data workflows","limit":3,"scrapeOptions":{"formats":["markdown"]}}'

A completed query costs two credits. Each successfully fetched result adds one credit. Three requested results with content therefore reserve up to five credits; failed page fetches do not add page credits. Search hits appear in data.web; content and source metadata are merged into hits whose fetches succeed.