There is a Chrome and Firefox extension called Characterizer that replaces RTK keywords with kanji on websites. I wrote a simple Ruby script that does the same thing for plain text files:

require "open-uri"
input = IO.read("input.txt")
open("http://jptxt.net/rtk-keywords.txt").read.split("\n").grep(/^[^#]/)[0..2199].each { |k|
kanji, keyword = k.split(";").values_at(0, 2)
input.gsub!(/(\b|[“”‘’…—])#{Regexp.escape(keyword)}(\b|[“”‘’…—])/i, "\\1#{kanji}\\2")
}
puts input

The output looks like this:

Jofuku was the Wise 男 之 China. 多 books 彼 読, and 彼 never
forgot 何 was 中 them. 皆 the characters 彼 knew as 彼 knew the lines
中 the palm 之 his 手. 彼 learned secrets 乃 birds and beasts, and
herbs and flowers and trees, and rocks and metals. 彼 knew magic and
poetry and 哲. 彼 grew 満 之 years and 智. 皆 the 民
honoured him; but 彼 was 勿 happy, for 彼 had a 語 written upon his
心.

If you use OS X, you can run the script by for example saving it as ~/script.rb, saving the input text as ~/input.txt, and then running "ruby script.rb" in Terminal.

I uploaded a sample output text to http://jptxt.net/japanese-fairy-tales.txt.

This seems more dangerous than helpful to me, because some of the characters are being used in totally different ways than in Japanese. It's also kanji->English studying, which is not recommended either by Heisig or most people on this forum.

For instance:
中 the palm 之 his 手. 彼 learned secrets 乃 birds and beasts

之 isn't used in modern Japanese outside of names, and it doesn't really map to "of". I don't even know what 乃 is supposed to mean here. "From"? 勿 is only "never" in old texts, mostly classical Chinese.

It just seems strange to study this way to me -- many of these phrases, in Japanese, would not use the kanji that appear in these script replacements, so you're not really studying Japanese with this, just RTK.

Such being the 州 之 affairs
州 means "state" in the sense of a division of a country, not state of affairs.

There was already an add-on for Firefox which did about the same:
http://kanjilish.mozdev.org/
I tried it some time ago but, I must say, never used it seriously for studying.
It may be useful if one uses it just for drilling the kanji-keyword associations, while keeping it completely separate from the actual meaning of the sentence in which it appears.