Wow! have you go tour work cut out for you.

Just my 2 cents worth:
1) gather a broad sample of pages from basic text similar to dictionary definiftions (ranging from words like apple through basic obscenities) as well as pages with pop ups, and standard graphical web pages.

2) have potential member/audience parents rate the pages from 1-suitable for all ages through 18-must be at least this old to look at this trash. And comment as to why it is not fit for a younger age group.

This will give you a basis (LOL) to see what might be suitable for each age group. Now comes the fun part - enforcement. Many words have acceptable use and meaning in the correct context, (ignore graphics for a moment), in another context they are totally unacceptable.

If you are rating sites, how do you prevent/react to changes in content? Once you over ride your automated systems to allow a site because certain words are used in an acceptable context, the page is allowed regardless of content changes.

An interesting thing I have run accross, is that filters stopping "bad word sites" will catch a bad word, then block the site. Here is where ads/popups can be interesting, say the word sex is banned for our discussion here. We go to the site and do our work (today's ad is for beer - not a forbidden word). The next time we go to the site the ad has a check box sex - M,F ( a banned word) the site does not come up. If the site is added to a blocked list, it will never be usable again until it is cleared by a moderator, if it is scanned each time processing cost goes up, but the user is frustrated because they can use the site, sometimes, and are left wondering why

Good luck
dageek