Early access is open for developers. Ask for a key
SourcedSearch[1]

Privacy policy and terms.

StatusDRAFT · 17 NOV 2026

Draft. This text is a plain-language draft written for early access. It has not been reviewed by a lawyer and will change before general release. Last edited 17 November 2026.

SourcedSearch is a web search API for AI agents. This page says what we collect, what we do with it, and the rules for using the service. We have tried to write it so you can read it in five minutes.

Privacy policy1

What we collect from this website

  • What you type into the early access form: your name, email, company or project, expected number of searches, and your note.
  • Basic server logs that any web host keeps: the time of a visit, the page asked for, and the visitor’s internet address.

This website sets no advertising cookies and has no tracking scripts. Fonts are loaded from Google Fonts, which means Google’s servers see the request for the font files.

What we collect when you use the API

  • The requests your program sends: the question, the filters, and the time.
  • Which API key sent the request, so we can apply limits and find faults.

We do not ask for, and do not want, personal details about your end users. Please do not put them in a question if you can avoid it.

What we do with it

  • Reply to your early access request.
  • Run the service, fix faults, and stop abuse.
  • Study which questions return poor results, so we can make results better.

We do not sell personal data. We do not use it for advertising.

How long we keep it

Form messages are kept for as long as we are in touch with you about early access. API request logs are kept for a short period and then deleted or stripped of the API key. The exact periods will be stated here before general release.

Your choices

You can ask to see, correct or delete what we hold about you. Write to hello@sourcedsearch.com. We answer every request.

If you own a website we have read2

Our crawler is the program that fetches public pages. It names itself in its requests and follows the rules in your robots.txt file. We store the text of public pages so that short passages can be returned with a link to the page.

To have your site or a page left out, block our crawler in robots.txt or write to hello@sourcedsearch.com from an address on your domain. We will remove it and confirm by email.

Terms of use3

The service

SourcedSearch returns passages from public web pages, with the source address and dates. It is in early access. It may change, slow down or be unavailable without notice. There is no uptime promise during early access.

Your account and key

Keep your API key secret. You are responsible for requests made with it. Tell us quickly if you think it has leaked and we will replace it.

Fair use

  • Do not use the service to break the law or to harm people.
  • Do not try to rebuild our index by copying results in bulk.
  • Do not remove the source from a passage and present the words as your own.
  • Stay within the request limits we give you.

The content belongs to its authors

The passages we return were written by other people and remain theirs. We give you no rights in that content. Quoting short passages with a link is your decision and your responsibility under the law where you operate.

No promise that a passage is true

We tell you what a page said and when. We do not check whether it is correct. Do not rely on the service alone for medical, legal, financial or safety decisions.

Price

Early access is free. If that changes, we will tell you by email before any charge, and you can stop using the service.

Liability

The service is provided as it is. To the extent the law allows, we are not liable for losses that come from using it during early access. This section will be rewritten with legal advice before general release.

Contact

Questions about any of this: hello@sourcedsearch.com.