A search engine reports that many thousands of pages contain the keywords you just entered. That is fine, but the first returned pages are just better advertised and probably not of big interest. Especially if you are looking for the best sources of information. On the other hand, it is hardly possible to check by clicking and viewing the reported thousands. The search engine will not allow you go too far, usually further then first 1000 , anyway. As a result, the most precious information remains hidden and not accessible. Sekhol Finder is designed to solve this problem. It combines web crawler functionality with advanced keywords text search and logical analysis of a document layout. The application starts with consulting main search engines and then proceeds downloading and processing pages using the links from the meaningful part of the relevant documents. The automatic processing dramatically improves efficiency of the search and allows to find the documents which are deep in the Internet. The search could be done in completely autonomous mode or be supervised and fine-tuned during the search. After most of the routine job is done by the application, the found pages are available for an instant offline preview, saving a lot of time for a user. The application creates summaries and also thumbnail images of the relevant documents. The final search results are put into the customisable auto generated html report, which is convenient for storing the links to the found pages or for creating a personal web links library.