<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:wfw="http://wellformedweb.org/CommentAPI/"
	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/"
	xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
	>

<channel>
	<title>stuff &#187; Security</title>
	<atom:link href="http://blog.beford.org/category/security/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.beford.org</link>
	<description></description>
	<lastBuildDate>Fri, 13 Aug 2010 22:59:52 +0000</lastBuildDate>
	<language>en</language>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.0.1</generator>
		<item>
		<title>Cosas que pasan</title>
		<link>http://blog.beford.org/2008/11/02/cosas-que-pasan/</link>
		<comments>http://blog.beford.org/2008/11/02/cosas-que-pasan/#comments</comments>
		<pubDate>Sun, 02 Nov 2008 06:56:25 +0000</pubDate>
		<dc:creator>Fernando</dc:creator>
				<category><![CDATA[Español]]></category>
		<category><![CDATA[Security]]></category>
		<category><![CDATA[XSS]]></category>

		<guid isPermaLink="false">http://blog.beford.org/?p=30</guid>
		<description><![CDATA[El mmg de Hades publico esto antes que yo, un XSS en live mail, estaba almacenado en /stuff hace tiempo ya, es un simple poc, que muestra los contactos.  Fue reportado a Microsoft hace casi 1 mes, recibi un correo diciendo que ya &#8230; <a href="http://blog.beford.org/2008/11/02/cosas-que-pasan/">Continue reading <span class="meta-nav">&#8594;</span></a>]]></description>
			<content:encoded><![CDATA[<p>El mmg de Hades publico esto antes que yo, un XSS en live mail, estaba almacenado en /stuff hace tiempo ya, es un simple poc, que muestra los contactos.  Fue reportado a Microsoft hace casi 1 mes, recibi un correo diciendo que ya habrian creado un reporte interno, donde se encargarian del asunto.</p>
<p>POC: <a href="http://beford.org/stuff/live.htm">http://beford.org/stuff/live.htm</a></p>
<p>Y otra cosa interesante que encontre, es algo en la <a href="http://m.google.com">interface movil de google</a>, en la seccion para configurar el idioma, el parametro continue, recibe la ruta donde se encontraba el usuario, para poder retornarlo alli una vez seleccione el idioma. Aparentemente no estan haciendo una concatenacion de 'm.google.com/' + continue sino de 'm.google.com' + continue. Pasandole a continue un valor de '.otrodominio.com' hariamos que todos los enlaces apunten a un sitio externo fuera de google.</p>
<p>POC: <a href="http://m.google.com/languages/?dc=gorganic&amp;continue=.beford.org">http://m.google.com/languages/?dc=gorganic&amp;continue=.beford.org</a></p>
]]></content:encoded>
			<wfw:commentRss>http://blog.beford.org/2008/11/02/cosas-que-pasan/feed/</wfw:commentRss>
		<slash:comments>7</slash:comments>
		</item>
		<item>
		<title>Congreso Nacional de Hacking Etico</title>
		<link>http://blog.beford.org/2008/05/08/congreso-nacional-de-hacking-etico/</link>
		<comments>http://blog.beford.org/2008/05/08/congreso-nacional-de-hacking-etico/#comments</comments>
		<pubDate>Thu, 08 May 2008 12:57:20 +0000</pubDate>
		<dc:creator>Fernando</dc:creator>
				<category><![CDATA[Español]]></category>
		<category><![CDATA[Security]]></category>
		<category><![CDATA[Uncategorized]]></category>

		<guid isPermaLink="false">http://blog.beford.org/?p=23</guid>
		<description><![CDATA[El dia de ayer paseando por el campus de mi Universidad, me he encontrado con este anuncio el cual me llamo la atencion: La Universidad Minuto de Dios ha organizado un Congreso de Hacking Etico en la ciudad de Bogota, &#8230; <a href="http://blog.beford.org/2008/05/08/congreso-nacional-de-hacking-etico/">Continue reading <span class="meta-nav">&#8594;</span></a>]]></description>
			<content:encoded><![CDATA[<p>El dia de ayer paseando por el campus de <a href="http://www.unal.edu.co/">mi Universidad</a>, me he encontrado con este anuncio el cual me llamo la atencion:</p>
<p style="text-align: center;"><a href="http://blog.beford.org/wp-content/uploads/2008/05/hacking.jpg"><img class="alignnone size-medium wp-image-24" title="hacking" src="http://blog.beford.org/wp-content/uploads/2008/05/hacking-225x300.jpg" alt="Poster Congreso Nacional Hacking Etico" width="121" height="163" /></a></p>
<p>La <a href="http://portal.uniminuto.edu/">Universidad Minuto de Dios</a> ha organizado un Congreso de Hacking Etico en la ciudad de Bogota, esta planeado para el Sabado 17 de Mayo, pueden ver toda la informacion sobre la conferencia en <a href="http://congresohacking.com/">http://congresohacking.com/</a> (al parecer algun inteligente lo reporto a Google como sitio de ataque, <a href="http://hacking.uniminuto.edu">http://hacking.uniminuto.edu</a>/). El precio de entrada es bastante accesible, la modica suma de 60.000 COP (33 USD) para los estudiantes universitarios, y 100.000 COP para los profesionales, asi que probablemente estare por alla, y si alguien quiera patrocinarme mejor! Tambien aparentemente habra un 'reto' hacking, sin saber que tipo de reto seran no me arriesgaria a inscribirme, dudo que sea algun concurso al estilo pwn2own, especialmente que los premios sean asi <img src='http://blog.beford.org/wp-includes/images/smilies/icon_razz.gif' alt=':P' class='wp-smiley' /> . Aca pueden ver el <a href="http://hacking.uniminuto.edu/images/programa.png">programa</a> de las conferencias. Al parecer muchos miembros de la comunidad de DragonJar estaran por alla.</p>
<p><a href="http://hacking.uniminuto.edu/images/programa.png"><br />
</a></p>
]]></content:encoded>
			<wfw:commentRss>http://blog.beford.org/2008/05/08/congreso-nacional-de-hacking-etico/feed/</wfw:commentRss>
		<slash:comments>0</slash:comments>
		</item>
		<item>
		<title>Gmail Cross Site Scripting</title>
		<link>http://blog.beford.org/2008/05/04/gmail-cross-site-scripting/</link>
		<comments>http://blog.beford.org/2008/05/04/gmail-cross-site-scripting/#comments</comments>
		<pubDate>Sun, 04 May 2008 01:40:13 +0000</pubDate>
		<dc:creator>Fernando</dc:creator>
				<category><![CDATA[Español]]></category>
		<category><![CDATA[Security]]></category>
		<category><![CDATA[XSS]]></category>
		<category><![CDATA[gecko]]></category>
		<category><![CDATA[gmail]]></category>
		<category><![CDATA[google]]></category>
		<category><![CDATA[opera]]></category>

		<guid isPermaLink="false">http://blog.beford.org/?p=19</guid>
		<description><![CDATA[El dia de hoy les traigo un pequeño descubrimiento, una vulnerabilidad en el sistema de Presentaciones (para los burros, powerpoint) en linea de Google Mail. El problema es sencillo, y no le tomara mucho tiempo arreglar a los de Mountain View, pero hay algo que me preocupa más. Describire el problema de Gmail rapidamente para poder pasar al detalle del plugin de flash. <a href="http://blog.beford.org/2008/05/04/gmail-cross-site-scripting/">Continue reading <span class="meta-nav">&#8594;</span></a>]]></description>
			<content:encoded><![CDATA[<p>El dia de hoy les traigo un pequeño descubrimiento, una vulnerabilidad en el sistema de Presentaciones (para los burros, powerpoint) en linea de Google Mail. El problema es sencillo, y no le tomara mucho tiempo arreglar a los de Mountain View, pero hay algo que me preocupa más. Describire el problema de Gmail rapidamente para poder pasar al detalle del plugin de flash.</p>
<p>Google Slideshows no esta filtrando los hipervinculos de las presentaciones cuyo destino es una URI javascript:, con solo crear un simple enlace, colocarle como destino <em>javascript:alert(1)</em> e insertarlo en la presentacion, es posible ejecutar javascript en el contexto de mail.google.com. Para los que quieran verlo en funcionamiento he preparado un sencillo poc que envia las cookies a un sitio remoto (que no existe), tendrian que enviarse <a href="http://beford.org/stuff/mozilla.ppt">este archivo</a> por correo, y luego verlo en modo de Presentacion desde Gmail.</p>
<p>Aqui es donde entra el detalle, estos enlaces son creados como objetos tipo flash, y son llamados para abrirse en una ventana nueva, mediante el parametro '_blank'. Aparentemente cada navegador maneja de manera distinta y son los culpables de este problema. Por ejemplo con este <a href="http://beford.org/stuff/geturl.swf">archivo</a>, al clickear el recuadro que enlaza a javascript:alert(document.domain), Firefox, aunque abre una nueva ventana con el contexto about:blank, internamente no actualiza el contexto de ejecucion por completo y es posible acceder a ciertos objetos del DOM (document.cookie, document.domain) pero no a otros (XMLHttpRequest). Opera por otra parte, crea la nueva ventana, pero al ejecutar la javascript uri, lo realiza en la misma ventana donde se origino el evento, comprometiendo completamente el contexto del sitio que hostea el objeto flash. En versiones anteriores de IE, el comportamiento era, abrir una nueva ventana, y el javascript se ejecutaba correctamente como about:blank, pero en las mas recientes, aparece una nueva ventana, y se cierra instantaneamente, sin poder saber que ocurre. En safari el comportamiento es distinto, el navegador aparentemente intenta cargar una pelicula al clickear el recuadro, y queda vacio, pero no ejecuta el javascript.</p>
<p><strong>Actualización</strong>: HaDeS preparo y publico un pequeño <a href="http://www.youtube.com/watch?v=6GANTR7tqu8">video en youtube</a> demostrando la falla, y Google ha deshabilitado temporalmente esta caracteristica mientras solucionan el problema.</p>
]]></content:encoded>
			<wfw:commentRss>http://blog.beford.org/2008/05/04/gmail-cross-site-scripting/feed/</wfw:commentRss>
		<slash:comments>0</slash:comments>
		</item>
		<item>
		<title>Security Bookmarklets</title>
		<link>http://blog.beford.org/2008/01/28/security-bookmarklets/</link>
		<comments>http://blog.beford.org/2008/01/28/security-bookmarklets/#comments</comments>
		<pubDate>Mon, 28 Jan 2008 23:47:57 +0000</pubDate>
		<dc:creator>Fernando</dc:creator>
				<category><![CDATA[English]]></category>
		<category><![CDATA[Programming]]></category>
		<category><![CDATA[Security]]></category>
		<category><![CDATA[Uncategorized]]></category>
		<category><![CDATA[firefox]]></category>
		<category><![CDATA[ie]]></category>

		<guid isPermaLink="false">http://blog.beford.org/?p=13</guid>
		<description><![CDATA[I took a couple of days from my school vacation to write 3 bookmarklets that will help me when auditing web sites, I'd like to share them here because I know that they'll help some of my friends, and probably &#8230; <a href="http://blog.beford.org/2008/01/28/security-bookmarklets/">Continue reading <span class="meta-nav">&#8594;</span></a>]]></description>
			<content:encoded><![CDATA[<p>I took a couple of days from my school vacation to write 3 bookmarklets that will help me when auditing web sites, I'd like to share them here because I know that they'll help some of my friends, and probably one of the two readers of this blog.</p>
<p><a class="bml" href="javascript:(function(){ x=prompt('Input text:','');l=x.length;t=x.charCodeAt(0);for (i = 1;i&lt;l;i++) t = t +','+x.charCodeAt(i);prompt('Output: ', 'Char('+t+')'); })();">Text2SQLChar</a> Converts an string into a CHAR() mysql, usefull when magic_quotes is on.<br />
<a class="bml" href="javascript:(function(){url=decodeURI(location);if (url.indexOf('select 1')==-1) { alert('Inject a simple \'union select 1\' first.');return; }max=url.substring(url.indexOf('select 1')).split(',').length;if (max&gt;1) zx = (max-1)+','+max; else zx = 'select 1';tmp=url.indexOf(zx);location=encodeURI(url.substring(0, tmp+zx.length)+','+(max+1)+url.substring(tmp+zx.length));})()">SQLIncrement</a> Increments automatically the number of columns of the injected select query.<br />
<a class="bml" href="javascript:(function(){url=decodeURI(location);if (url.indexOf('select 1')==-1) { alert('Inject a simple \'union select 1\' first.');return; }max=url.substring(url.indexOf('select 1')).split(',').length;if (max&gt;2) zx = (max-2)+','+(max-1); else { if (max==2) zx='select 1'; else return; } tmp=url.indexOf(zx);dx = ','+max;location=encodeURI(url.substring(0, tmp+zx.length)+url.substring(tmp+zx.length+dx.length));})()">SQLDecrement</a> Decrements automatically the number of columns of the injected select query.<br />
<a class="bml" href="javascript:(function(){ var e,s; IB=1; function isDigit(c) { return ('0' &lt;= c &amp;&amp; c &lt;= '9') } L = decodeURI(location.href); LL = L.length; for (e=LL-1; e&gt;=0; --e) if (isDigit(L.charAt(e))) { for(s=e-1; s&gt;=0; --s) if (!isDigit(L.charAt(s))) break; break; } ++s; if (e&lt;0) return; oldNum = L.substring(s,e+1); newNum = '' + (parseInt(oldNum,10) + IB); while (newNum.length &lt; oldNum.length) newNum = '0' + newNum; location.href = L.substring(0,s) + newNum + L.slice(e+1); })();">Increment</a> Allows you to navigate up html files or images that have a number in them.<br />
<a class="bml" href="javascript:(function(){ var e,s; IB=-1; function isDigit(c) { return ('0' &lt;= c &amp;&amp; c &lt;= '9') } L = decodeURI(location.href); LL = L.length; for (e=LL-1; e&gt;=0; --e) if (isDigit(L.charAt(e))) { for(s=e-1; s&gt;=0; --s) if (!isDigit(L.charAt(s))) break; break; } ++s; if (e&lt;0) return; oldNum = L.substring(s,e+1); newNum = '' + (parseInt(oldNum,10) + IB); while (newNum.length &lt; oldNum.length) newNum = '0' + newNum; location.href = L.substring(0,s) + newNum + L.slice(e+1); })();">Decrement</a> Allows you to navigate down html files or images that have a number in them.<br />
<a class="bml" href="javascript:(function(){x=prompt('Text:','');l=x.length%3;if(l)for(i=1;i&lt;7-l;i++)x=x+'%20';;prompt('Output:',window.btoa(x));})();">base64</a> Firefox only, base64 with no padding</p>
<p>I wrote the first 3 of them, I'm not javascript god, so they need to get improved, I'll try to make them shorter, and follow a couple of suggestion from <a href="http://www.bookmarklets.com/about/make/js.html">bookmarklets.com</a>. The two others come from <a href="http://ha.ckers.org/bookmarklets.html">RSnake' bookmarklet collection</a>, where I fixed a detail, it was not decoding the URL before using it, so if the url contained an escaped value at the end of the url like %20, it would modify it and change it to %21. I'll keep this post updated whenever I find or write a new security-related bookmarklet. As another issue, I noticed that the site looks ugly on low resolution systems, so I'll try to get a new theme to fix that on a couple of days and update the wordpress version.</p>
]]></content:encoded>
			<wfw:commentRss>http://blog.beford.org/2008/01/28/security-bookmarklets/feed/</wfw:commentRss>
		<slash:comments>0</slash:comments>
		</item>
		<item>
		<title>Firefox 2.0.0.10 released</title>
		<link>http://blog.beford.org/2007/11/27/firefox-20010-released/</link>
		<comments>http://blog.beford.org/2007/11/27/firefox-20010-released/#comments</comments>
		<pubDate>Tue, 27 Nov 2007 05:40:52 +0000</pubDate>
		<dc:creator>Fernando</dc:creator>
				<category><![CDATA[English]]></category>
		<category><![CDATA[Security]]></category>
		<category><![CDATA[XSS]]></category>
		<category><![CDATA[firefox]]></category>
		<category><![CDATA[gecko]]></category>
		<category><![CDATA[jar]]></category>
		<category><![CDATA[mozilla]]></category>

		<guid isPermaLink="false">http://blog.beford.org/?p=11</guid>
		<description><![CDATA[Mozilla Corporation just released Firefox 2.0.0.10 which includes fixes against JAR uri attacks. This issue affected browsers that used Gecko engine, a quick check showed me that only K-meleon browser was also updated, however there are several Gecko based web &#8230; <a href="http://blog.beford.org/2007/11/27/firefox-20010-released/">Continue reading <span class="meta-nav">&#8594;</span></a>]]></description>
			<content:encoded><![CDATA[<p>Mozilla Corporation just released <a href="http://www.mozilla-europe.org/es/products/firefox/2.0.0.10/releasenotes/">Firefox 2.0.0.10</a> which includes fixes against <a href="http://blog.beford.org/?p=8">JAR uri attacks</a>. This issue affected browsers that used Gecko engine, a quick check showed me that only <a href="http://kmeleon.sourceforge.net/">K-meleon</a> browser was also updated, however there are several Gecko based web browsers that need to get fixed: <a href="http://en.wikipedia.org/wiki/List_of_web_browsers#Gecko-based_browsers">Gecko-based browsers.</a></p>
<p><strong>Update: </strong>Let's make that, Firefox 2.0.0.11, which also fixes some regressions.</p>
]]></content:encoded>
			<wfw:commentRss>http://blog.beford.org/2007/11/27/firefox-20010-released/feed/</wfw:commentRss>
		<slash:comments>0</slash:comments>
		</item>
		<item>
		<title>Google Gadgets XSS (IE6/Opera)</title>
		<link>http://blog.beford.org/2007/11/25/google-gadgets-xss-ie6/</link>
		<comments>http://blog.beford.org/2007/11/25/google-gadgets-xss-ie6/#comments</comments>
		<pubDate>Sun, 25 Nov 2007 09:58:23 +0000</pubDate>
		<dc:creator>Fernando</dc:creator>
				<category><![CDATA[English]]></category>
		<category><![CDATA[Security]]></category>
		<category><![CDATA[XSS]]></category>
		<category><![CDATA[google]]></category>
		<category><![CDATA[ie]]></category>
		<category><![CDATA[xml]]></category>

		<guid isPermaLink="false">http://blog.beford.org/?p=9</guid>
		<description><![CDATA[There is a recent discussion on ha.ckers.org regarding a possible CSRF that could allow an attacker to inject an evil gadget on someobdy else's iGoogle page. After checking the format of the xml file used to define the gadgets properties, &#8230; <a href="http://blog.beford.org/2007/11/25/google-gadgets-xss-ie6/">Continue reading <span class="meta-nav">&#8594;</span></a>]]></description>
			<content:encoded><![CDATA[<p>There is a recent discussion on <a href="http://ha.ckers.org/blog/20071119/google-gadgets-gaffe/">ha.ckers.org</a> regarding a possible CSRF that could allow an attacker to inject an evil gadget on someobdy else's iGoogle page. After checking the format of the <a href="http://ha.ckers.org/asdf2.xml">xml file</a> used to define the gadgets properties, I noticed a couple of attributes that could be used as point of injections to active content, the <strong>thumbnail</strong> and <strong>screenshot </strong>attribute. Only one of them is vulnerable, the screenshot attribute, by using a javascript URI as value you can execute active content on certain browsers such as IE6. This is a poc that shows an alert with the current document.domain value:</p>
<p><a href="http://www.google.com/ig/adde?moduleurl=http://beford.org/stuff/ig.xml">www.google.com/ig/adde?moduleurl=http://beford.org/stuff/ig.xml</a></p>
<p>I've just notified Google about this, to avoid been a target of this flaw you can switch to Mozilla Firefox (+Noscript), or upgrade to IE7, which is not an option for Windows 2000 users.</p>
<p><strong>Update:</strong> According to a friend, Rafael, Opera users are vulnerable aswell, he sent me an screenshot which you can see here: <a href="http://blog.beford.org/wp-content/uploads/2007/11/ss.jpg" title="Opera Google XSS">Opera Google XSS.</a></p>
<p><strong>Update 2: </strong>Google fixed this on 27 November.</p>
]]></content:encoded>
			<wfw:commentRss>http://blog.beford.org/2007/11/25/google-gadgets-xss-ie6/feed/</wfw:commentRss>
		<slash:comments>2</slash:comments>
		</item>
		<item>
		<title>Firefox jar: Protocol Vulnerability</title>
		<link>http://blog.beford.org/2007/11/10/firefox-jar-protocol-vulnerability/</link>
		<comments>http://blog.beford.org/2007/11/10/firefox-jar-protocol-vulnerability/#comments</comments>
		<pubDate>Sat, 10 Nov 2007 10:47:45 +0000</pubDate>
		<dc:creator>Fernando</dc:creator>
				<category><![CDATA[English]]></category>
		<category><![CDATA[Security]]></category>
		<category><![CDATA[XSS]]></category>
		<category><![CDATA[firefox]]></category>
		<category><![CDATA[gecko]]></category>
		<category><![CDATA[google]]></category>
		<category><![CDATA[jar]]></category>
		<category><![CDATA[mozilla]]></category>

		<guid isPermaLink="false">http://blog.beford.org/?p=8</guid>
		<description><![CDATA[I just came across pdp's finding jar protocol vulnerability on Mozilla Firefox, I think its a big issue, and the fact that it has been on bugzilla (#369814) for way more than ten fuck*ng days is not a good thing. &#8230; <a href="http://blog.beford.org/2007/11/10/firefox-jar-protocol-vulnerability/">Continue reading <span class="meta-nav">&#8594;</span></a>]]></description>
			<content:encoded><![CDATA[<p>I just came across pdp's finding <a href="http://www.gnucitizen.org/blog/web-mayhem-firefoxs-jar-protocol-issues"> jar protocol vulnerability</a> on Mozilla Firefox, I think its a big issue, and the fact that it has been on <a href="https://bugzilla.mozilla.org/show_bug.cgi?id=369814">bugzilla (#369814)</a> for way more than <a href="http://ha.ckers.org/blog/20070803/mozilla-says-ten-fucking-days/">ten fuck*ng days</a> is not a good thing.</p>
<p>According to pdp, this issue makes vulnerable to Cross-site scripting applications that allow users uploading compressed ZIP, and JAR files. After a couple of minutes messing around the poc's, I figured out that sites with open redirect issues are vulnerable too. I've created this poc that attacks Gmail, it's based on my previous post and it will only show your contacts list, it's not being logged server side or anything (as some people thought that my previous poc did. Credit to <a href="http://lowtechlive.com/">tx</a> for discovering the <a href="http://sla.ckers.org/forum/read.php?3,505,10958#msg-10958">open redirect issue</a> used to exploit Google / Firefox):</p>
<p><a href="http://beford.org/stuff/jarjarbinks.htm">http://beford.org/stuff/jarjarbinks.htm</a></p>
<p>Who's fault? Both, <a href="http://ha.ckers.org/blog/20070520/phishing-through-google-yet-again/">Google</a> <a href="http://ha.ckers.org/blog/20070224/google-fixes-one-redirect-but-leaves-lots-of-others/">for</a> <a href="http://ha.ckers.org/blog/20061224/google-redirects-help-phishers-again/">having</a> <a href="http://ha.ckers.org/blog/20060822/google-redirection-hole-used-for-phishing/">open</a> <a href="http://sla.ckers.org/forum/read.php?3,505,page=3#msg-1349">redirect</a> issues and not fixing them, and Mozilla Corporation for failing to address this problem.</p>
<p>What can I do to protect myself? Giorgio Maone have already added protection against this flaw to <a href="http://noscript.net/getit#devel">NoScript development version</a>.</p>
<p><strong> Update: </strong>NoScript released stable version with Jar protection.  A new <a href="https://bugzilla.mozilla.org/show_bug.cgi?id=403331">bugzilla (#403331) </a>entry was created to fix the inappropiate redirect on jar protocol, according to the lastest comments and bug keyword, there seems to be a patch and will be availible with Firefox 2.0.0.10.</p>
]]></content:encoded>
			<wfw:commentRss>http://blog.beford.org/2007/11/10/firefox-jar-protocol-vulnerability/feed/</wfw:commentRss>
		<slash:comments>21</slash:comments>
		</item>
		<item>
		<title>Google Vulnerability</title>
		<link>http://blog.beford.org/2007/09/24/googlecom-cross-site-scripting-vulnerability/</link>
		<comments>http://blog.beford.org/2007/09/24/googlecom-cross-site-scripting-vulnerability/#comments</comments>
		<pubDate>Mon, 24 Sep 2007 03:29:26 +0000</pubDate>
		<dc:creator>Fernando</dc:creator>
				<category><![CDATA[English]]></category>
		<category><![CDATA[Security]]></category>
		<category><![CDATA[XSS]]></category>
		<category><![CDATA[google]]></category>
		<category><![CDATA[jar]]></category>

		<guid isPermaLink="false">http://blog.beford.org/?p=3</guid>
		<description><![CDATA[Yesterday, I found a new Google.com XSS vulnerability that can be abused to steal information from Gmail accounts, I've done responsible disclosure of at least 3 vulns to Google, but since I haven't got enough 'motivation', I'll go full disclosure &#8230; <a href="http://blog.beford.org/2007/09/24/googlecom-cross-site-scripting-vulnerability/">Continue reading <span class="meta-nav">&#8594;</span></a>]]></description>
			<content:encoded><![CDATA[<p>Yesterday, I found a new Google.com XSS vulnerability that can be abused to steal information from Gmail accounts, I've done responsible disclosure of at least 3 vulns to Google, but since I haven't got enough 'motivation', I'll go full disclosure now.</p>
<p>The vulnerability exists in Blogspot polls feature, I had already disclosed a vulnerability on this system. The 'font' parameter was not being sanitized before being used inside an STYLE tag, so you could inject IE's expression() and Mozilla's -moz-binding. They fixed it, however they didn't check enough the rest of the code, the new XSS is:</p>
<p><a href="http://www.google.com/reviews/polls/display/159769971366811755/blogger_template/vote?purl=blogspot.com%2F&amp;chrtclr=%23599be2&amp;hideq=false&amp;font=normal+normal+100%25+Helvetica%2CArial%2CVerdana%2C%20Trebuchet+MS%20%2C+Sans-serif&amp;u_tz=%22%3EXSS">Simple XSS POC</a></p>
<p>Since its Sunday and there is nothing else to do, I've created 2 more pocs, one of them, shows a your contacts, the  second one will make Gmail forward all new received emails to another email account,  no user interaction required, well you just need to open a website while still logged to Gmail.</p>
<p>POC1: <a href="http://beford.org/stuff/contacts.htm">http://beford.org/stuff/contacts.htm</a><br />
POC2: <a href="http://beford.org/stuff/gmail.htm">http://beford.org/stuff/gmail.htm</a></p>
<p>I have tested them under IE, Konqueror, Opera and Firefox, it should work on all of them. If you want to be protected against this kind of attacks, I'd highly recommend Firefox + <a href="http://noscript.net/">NoScript</a>.</p>
<p><strong>Update</strong>: Google fixed this issue, I'd like to ask the people that looked at the second poc to disable forwarding if you have not done so, I'm still getting ton of email. This screenshot shows how to disable forwarding.</p>
<p><a href="http://blog.beford.org/wp-content/uploads/2007/09/gmail.png" title="gmail.png"><img src="http://blog.beford.org/wp-content/uploads/2007/09/gmail.thumbnail.png" alt="gmail.png" /></a></p>
]]></content:encoded>
			<wfw:commentRss>http://blog.beford.org/2007/09/24/googlecom-cross-site-scripting-vulnerability/feed/</wfw:commentRss>
		<slash:comments>46</slash:comments>
		</item>
	</channel>
</rss>
