vim 6.0 and 6.1, and possibly other versions, allows attackers to execute arbitrary commands using the libcall feature in modelines, which are not sandboxed but may be executed when vim is used to edit a malicious file, as demonstrated using mutt.
https://exchange.xforce.ibmcloud.com/vulnerabilities/10835
http://www.securityfocus.com/bid/6384
http://www.redhat.com/support/errata/RHSA-2002-302.html
http://www.redhat.com/support/errata/RHSA-2002-297.html
http://www.mandrakesoft.com/security/advisories?name=MDKSA-2003:012
http://www.guninski.com/vim1.html
http://sunsolve.sun.com/pub-cgi/retrieve.pl?doc=fsalert/55700
http://marc.info/?l=bugtraq&m=108077992208690&w=2
http://lists.grok.org.uk/pipermail/full-disclosure/2002-December/002948.html
http://distro.conectiva.com.br/atualizacoes/?id=a&anuncio=000812