By linking to a specific url in Plone 2.5-5.1rc1 with a parameter, an attacker could send you to his own website. On its own this is not so bad: the attacker could more easily link directly to his own website instead. But in combination with another attack, you could be sent to the Plone login form and login, then get redirected to the specific url, and then get a second redirect to the attacker website. (The specific url can be seen by inspecting the hotfix code, but we don't want to make it too easy for attackers by spelling it out here.)

When you visit a page where you need to login, Plone 2.5-5.1rc1 sends you to the login form with a 'came_from' parameter set to the previous url. After you login, you get redirected to the page you tried to view before. An attacker might try to abuse this by letting you click on a specially crafted link. You would login, and get redirected to the site of the attacker, letting you think that you are still on the original Plone site. Or some javascript of the attacker could be executed. Most of these types of attacks are already blocked by Plone, using the `isURLInPortal` check to make sure we only redirect to a page on the same Plone site. But a few more ways of tricking Plone into accepting a malicious link were discovered, and fixed with this hotfix.

Accessing private content via str.format in through-the-web templates and scripts in Plone 2.5-5.1rc1. This improves an earlier hotfix. Since the format method was introduced in Python 2.6, this part of the hotfix is only relevant for Plone 4 and 5.

Cross-site scripting (XSS) vulnerability in the manage_findResult component in the search feature in Zope ZMI in Plone before 4.3.12 and 5.x before 5.0.7 allows remote attackers to inject arbitrary web script or HTML via vectors involving double quotes, as demonstrated by the obj_ids:tokens parameter. NOTE: this vulnerability exists because of an incomplete fix for CVE-2016-7140.

Zope before 2.13.19, as used in Plone before 4.2.3 and 4.3 before beta 1, does not reseed the pseudo-random number generator (PRNG), which makes it easier for remote attackers to guess the value via unspecified vectors. NOTE: this issue was SPLIT from CVE-2012-5508 due to different vulnerability types (ADT2).

The error pages in Plone before 4.2.3 and 4.3 before beta 1 allow remote attackers to obtain random numbers and derive the PRNG state for password resets via unspecified vectors. NOTE: this identifier was SPLIT per ADT2 due to different vulnerability types. CVE-2012-6661 was assigned for the PRNG reseeding issue in Zope.

The batch id change script (renameObjectsByPaths.py) in Plone before 4.2.3 and 4.3 before beta 1 allows remote attackers to change the titles of content items by leveraging a valid CSRF token in a crafted request.

registerConfiglet.py in Plone before 4.2.3 and 4.3 before beta 1 allows remote attackers to execute Python code via unspecified vectors, related to the admin interface.

Cross-site scripting (XSS) vulnerability in kssdevel.py in Plone before 4.2.3 and 4.3 before beta 1 allows remote attackers to inject arbitrary web script or HTML via unspecified vectors.

gtbn.py in Plone before 4.2.3 and 4.3 before beta 1 allows remote authenticated users with certain permissions to bypass the Python sandbox and execute arbitrary Python code via unspecified vectors.