A quick update on the state of VulnPryer. After a couple of quiet months, activity has picked back up on this project to improve our vulnerability prioritization efforts. We’re close to wrapping up a long-delayed effort to move the generation of our customized scoring algorithm from a manual Vagrant-driven process to a fully automated and scheduled process running on AWS using Data Pipelines and OpsWorks. At this point we are pulling our data feeds, re-scoring thousands of vulnerabilities through our draft algorithm, and generating a revised data file for downstream ingestion and risk analysis on a daily basis.

While a full post will be forthcoming, if you’re tracking the Chef-VulnPryer repository you’ll see a few updates as some of the kinks are worked out in the supporting infrastructure (flipping between AWS Linux and Ubuntu images is…interesting). I’m excited to see this work completed and to move out of the perennial data wrangling aspects and move back into the modelling and algo testing.

Stay tuned!