<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: On CSRF</title>
	<atom:link href="http://www.greebo.net/2007/04/02/on-csrf/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.greebo.net/2007/04/02/on-csrf/</link>
	<description>mostly useless crap from me</description>
	<lastBuildDate>Sun, 21 Feb 2010 00:51:44 +0000</lastBuildDate>
	<generator>http://wordpress.org/?v=2.9</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: vanderaj</title>
		<link>http://www.greebo.net/2007/04/02/on-csrf/comment-page-1/#comment-7945</link>
		<dc:creator>vanderaj</dc:creator>
		<pubDate>Mon, 23 Apr 2007 00:42:33 +0000</pubDate>
		<guid isPermaLink="false">http://www.greebo.net/?p=409#comment-7945</guid>
		<description>Louis, 

you&#039;re right. After talking this potential method over with some folks, it&#039;s not that much better than a random token; knowing a secret doesn&#039;t necessarily mean you can&#039;t force the Javascript to execute things via CSRF directed at the existing client-side JS. 

I still think a positional token would hamper all but the cleverest attacks, but this positional token doesn&#039;t require PKI, it just requires a random value. That would be of the same level of difficulty to attack as before.

Andrew</description>
		<content:encoded><![CDATA[<p>Louis, </p>
<p>you&#8217;re right. After talking this potential method over with some folks, it&#8217;s not that much better than a random token; knowing a secret doesn&#8217;t necessarily mean you can&#8217;t force the Javascript to execute things via CSRF directed at the existing client-side JS. </p>
<p>I still think a positional token would hamper all but the cleverest attacks, but this positional token doesn&#8217;t require PKI, it just requires a random value. That would be of the same level of difficulty to attack as before.</p>
<p>Andrew</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Louis van Ekert</title>
		<link>http://www.greebo.net/2007/04/02/on-csrf/comment-page-1/#comment-7827</link>
		<dc:creator>Louis van Ekert</dc:creator>
		<pubDate>Fri, 20 Apr 2007 00:34:32 +0000</pubDate>
		<guid isPermaLink="false">http://www.greebo.net/?p=409#comment-7827</guid>
		<description>Hi Andrew, how are things? - I read your diary with interest and irregularly.  
Just a Q. re: rsToken = sign(serverPublicKey, destination Flow ID, Flow Step);
are you implying that serverPublicKey is the signing key?  If so, anyone (incl. attack zombies) could sign.  Wouldn&#039;t you need some sort of shared secret to sign - like it is done in HMAC? 
All the best.
Louis</description>
		<content:encoded><![CDATA[<p>Hi Andrew, how are things? &#8211; I read your diary with interest and irregularly.<br />
Just a Q. re: rsToken = sign(serverPublicKey, destination Flow ID, Flow Step);<br />
are you implying that serverPublicKey is the signing key?  If so, anyone (incl. attack zombies) could sign.  Wouldn&#8217;t you need some sort of shared secret to sign &#8211; like it is done in HMAC?<br />
All the best.<br />
Louis</p>
]]></content:encoded>
	</item>
</channel>
</rss>
