I recall reading a comment saying that Lift deals well with most of
the Top 10 OWASP vulnerabilities (http://www.owasp.org/index.php/
Top_10_2007), but how?  Most, not all?  Are there strategies that are
recommended to deal with the weaknesses?

I am very interested in a Wiki post that explains how Lift addresses
the top 10 vulnerabilities.  (No pressure, but I needed this
yesterday.)  Please point me to any existing resource/documentation
that explains this.

Very often, I see points that mention how secure Lift is, and I
certainly appreciate the fact that Lift promotes security as a core
initiative.  I can also see how secure it is compared to other
strategies (yeah, you heard me ColdFusion!).  Keep up the good work.

Does anybody have experience submitting Lift Apps to third party
security assessment or vulnerability analysis?  Any tips from that
experience?

   1. Lift is resistant to XSS attacks.  By default pages are composed in
   XML rather than Strings.  It takes the developer extra work to insert XSS
   strings into output rather than having to make sure each String is properly
   escaped before being cat'ed to the output.
   2. Lift is resistant to SQL Injection attacks because mapper and JPA do
   not compose Strings into SQL statements, but rather bind well typed
   parameters into prepared statements.  So, if you go the normal path, you get
   SQL injection resistance.  If you want to manually craft a String to send as
   a query, in mapper you have to "sign" the string with the time, date and a
   certification that you've reviewed the String for SQL Injection problems.
   3. Lift never shells out.  You can't cause a file to be executed from a
   Lift app unless your app manually uses Java's Runtime.execute() call.
   4. By default, Lift creates opaque GUIDs to refer to components on the
   server side (whether that's a function to execute when a form field is
   submitted, what to do on an Ajax call, etc.)  By default, it's easier to use
   this callback mechanism than advertise a primary key or other sensitive
   piece of information.  Lift also has the KeyObfuscator which will create a
   session-specific mapping of primary keys to opaque ids.  Using
   KeyObfuscator, you can send JSON objects to the client with stable primary
   keys that are obfuscated and not usable outside the current session.
   5. By default Lift's form fields contain GUIDs that are cryptographically
   impossible to predict.  It's not possible to do CSRF because one does not
   know the name of form fields (they are not stable)
   6. Lift has different production vs. development mode error messages.
   There's little information that leaks about underlying configurations, even
   exceptions, in production mode.
   7. Lift uses the container's session management (usually JSESSIONID) for
   session management.  As far as I know, Jetty, Tomcat, Glassfish are secure
   in terms of the way they deal with sessions.  Of course, anything that's not
   over SSL is vulnerable to a cookie stealing attack.
   8. Crypto key storage is a container-level issue.  See #7
   9. See #7
   10. Lift's sitemap is the best and most secure integration of UI and
   page-level access.  You can look at the sitemap to determine the access
   control rules for a given page (it's declarative) and it's enforced long
   before your page gets accessed.

Hope this helps.