HackBlue

Tools · Ethical OSINT and privacy self-defense · 03 / 17

Using search engine operators to find your organization’s accidental exposure

Combine site:, filetype:, inurl:, and intitle: to find documents, login pages, and directory listings your own domain exposes, and get them removed or protected.

9 min read · practice only with authorization

Operators that matter

site:example.org restricts results to a domain and its subdomains. filetype:pdf or ext:xlsx finds documents. inurl:admin and intitle:"index of" find admin panels and directory listings. Quotes force exact phrases, and a minus sign excludes terms. Support varies by search engine, so try more than one.

Snippet
site:example.org filetype:pdf "internal use only"
site:example.org ext:xlsx OR ext:csv
site:example.org intitle:"index of"
site:example.org inurl:login OR inurl:admin -www
site:*.example.org -site:www.example.org

What to look for

Documents marked internal or confidential, spreadsheets with names and emails, backup or log files, test and staging sites, open directory listings, and login portals that should not be public. Each finding is a candidate for removal, access control, or at least awareness.

Lab: review your own domain

For your own personal domain or an organization domain you are authorized to review, run the queries above and record every unexpected result. For each, decide the fix: remove the file, require authentication, add the path to robots.txt only as a hint (robots.txt does not protect anything), or request removal from search results after the content is gone using the search engine’s removal tool.

Prevent it

Keep internal documents in authenticated systems, disable directory listing on web servers, add noindex headers to staging sites and protect them with authentication, and include a search-exposure review in your regular external assessment.

Snippet
# nginx: no directory listings, noindex on staging
autoindex off;
add_header X-Robots-Tag "noindex, nofollow" always;

Diagram

Search operators

Combine them for precise queries.

  • site:

    Limit to a domain or subdomain.

  • filetype: / ext:

    Limit to a document type.

  • intitle:

    Words in the page title, such as "index of".

  • inurl:

    Words in the URL, such as admin or login.

  • "exact" and -term

    Exact phrase; exclude a term.

Check yourself

1. Which operator limits results to one domain?

2. What does intitle:"index of" typically reveal?

3. Why is robots.txt not a fix for sensitive files?