Hi there,
The requirements are quite clear and straightforward to implement so no questions really. The crawler will visit each website, possibly disguised as google bot to circumvent anti-scraping precautions. That'll also speed up each request slightly since the response will be fully text. Then the crawler will search the response text for the keywords and find the existing keywords on that domain.
Since nothing will be hard coded into the crawler you can easily modify your lists, both domains and keywords, in the configuration file and run it again without any hassle.
I am planning to use python 3 branch to implement the crawler. You can simply start the console/cmd to run the crawler. If you have something else in mind, like if you want an actual GUI, please let me know then we can discuss other options.
I expect it to be ready in 5 days, thanks.