Google on Monday announced that it has made available its Sandboxed API as open source in an effort to make it easier for software developers to create secure products.
It’s not uncommon for applications to be affected by memory corruption or other types of vulnerabilities that can be exploited for remote code execution and other purposes. Using a sandbox ensures that the code responsible for processing user input can only access the resources it needs to, which mitigates the impact of a flaw by containing the exploit to a restricted environment and preventing it from interacting with other software components.
While sandboxing can be highly useful, Google says it’s often not easy to implement. That is why the internet giant has decided to open source its Sandboxed API, which should make it easier to sandbox C and C++ libraries. The company has also open sourced its core sandboxing project, Sandbox2, which can be used on its own to secure Linux processes.
“Sandboxed API makes it possible to create security policies for individual software libraries. This concept allows to create reusable and secure implementations of functionality residing within popular software libraries, yet is granular enough to protect the rest of used software infrastructure,” explained members of Google’s ISE Sandboxing team.
For the time being, the Sandboxed API only supports libraries written in C and only Linux. However, Google plans on adding support for more programming runtimes and more operating systems — Unix-like systems may be added sooner, but the company says porting it to Windows is more challenging.
The API’s developers also plan on adding new sandboxing technologies, including hardware virtualization, and extending the build system.
Google has also pointed out that anyone can make suggestions for improving the Sandboxed API, which is also covered by the company’s Patch Reward Program.
Related: Google Open Sources Fuzzing Platform