An example project that exploits the default typing issue in Jackson-databind (https://github.com/FasterXML/jackson-databind) via Spring application contexts and expressions
The Jackson-databind project has a feature called default-typing (not enabled by default). When the target class has some polymorph fields inside (such as interfaces, abstract classes or the Object base class), the library can include type info into the JSON structure and use that info at unmarshalling. This can be dangerous when the input is controlled by an attacker and the target class contains a field of type Object or something general (like Comparable).
How likely is this? I'm naive, so I hope Java developers don't degrade a type-safe language to the level of an interpreted type-unsafe language by (ab)using Objects as base classes... But I wouldn't be surprised if one day some huge enterprise software would be exploited one day via this vulnerability.
After the original discoveries (CVE-2017-7525) had been reported, the author patched this attack surface with a blacklist, which was incomplete (as by nature of blacklists). This proof-of-concept project is a follow-up to demonstrate one more way of exploitation; by abusing Spring classes via Jackson, this could lead to remote code execution. Note: FileSystemXmlApplicationContext is happy to fetch the specified Spring context from anywhere, even from remote location via http.
MITRE assigned CVE-2017-17485 to this vulnerability.
The following ones (inclusive) and older: 2.9.3, 2.7.9.1, 2.8.10
The fixed versions 2.7.9.2, 2.8.11 and 2.9.3.1 (which is to be released at time of writing these lines) expanded the blacklist once again so that Spring application contexts cannot be instantiated anymore.
The new major version (3.x) of Jackson-databind will address this topic via a new API layer that provides a way to achieve whitelisting-based serialization for these polymorph classes.
https://medium.com/@cowtowncoder/on-jackson-cves-dont-panic-here-is-what-you-need-to-know-54cd0d6e8062 https://adamcaudill.com/2017/10/04/exploiting-jackson-rce-cve-2017-7525/