Basic Information

appmixer.ai.exa.FindSearchResults

Search Web

1.0.0

Public

Run a semantic or keyword web search with Exa and return the matching pages, optionally with their cleaned page content.

Input Ports

in

FieldTypeDescription
Query text What to search the web for. Exa is semantic, so a descriptive phrase works better than keywords.
Search Type select Auto lets Exa pick the best method. Fast and Instant are cheaper and quicker, the Deep modes search harder and cost more per call - Deep (reasoning) adds stronger reasoning for complex analysis. These are the six modes Exa documents for /search.
Number Of Results number How many results to return, between 1 and 100.
Category select Restrict the search to one kind of page. Leave empty to search everything.
Include Domains text Comma separated list of domains to restrict the search to, e.g. wikipedia.org, bbc.co.uk.
Exclude Domains text Comma separated list of domains to leave out of the results.
Published After date-time Only return pages published on or after this date.
Published Before date-time Only return pages published on or before this date.
User Location text Two letter ISO country code of the searcher, e.g. US. Biases results towards that country.
Include Page Text toggle Also return the cleaned full text of each result. Costs extra Exa credits per page.
Include Highlights toggle Also return the snippets of each page that are most relevant to the query.
Summary Query text Ask Exa to summarise each result with this question in mind, e.g. What does this company sell?. Leave empty for no summary.
Output Type select Choose whether you want to receive the result set as one complete list, or first item only or one item at a time or stream the items to a file.

Output Ports

out

notFound

FieldTypeDescription
Query string -

Full Metadata