Data

The full and up-to-date dataset used on this site can be downloaded here. It is updated weekly and all the pages and data on this site are extracted from it. It is a JSON file as produced by the sec-certs cc command from the tool. Documentation for the tool, including a quick-start guide, can be found here.

The dataset was last successfully updated at 12:00 on 02.06.2025 using version 0.2.1.post1.dev34+g5f893fb of the sec-certs tool and it contains 6326 certificates.

Main dataset The main dataset is a JSON file with data on the certificates extracted from the certification artifacts (i.e. PDFs). It does not contain the files themselves or information about maintenance updates.
Maintenance updates The maintenance update dataset is a JSON file almost identical to the main dataset, except it covers only maintenance updates.
Scheme data The scheme dataset is a JSON file with data on certified products, products under evaluation and archived certificates scraped from the websites of individual Common Criteria national schemes.
Full datasset The full dataset contains all the datasets mentioned above as well as the certification artifacts themselves (i.e. PDFs). It is a gzip-compressed tar archive. Download this if you want to do your own processing on top of the data provided by our tooling.
sec-certs is also a library! You can easily work with these datasets using our Python library. Check out the docs.

Data sources

We are getting information from the following data sources:

  • The commoncriteriaportal.org, which we use as the trusted source of certificate data. We scrape both its HTML site and the CSV export of certificates and download all of the available PDFs: certificates, certification reports and security targets. We run a caching proxy to minimize the load on the CC portal and disperse it over time by keeping the cache hot.
  • The sites of national schemes, which we use to enrich the certificate metadata. We try to scrape all of the scheme sites for a list of certified products, though not all publish this list properly. For some, we also scrape a list of products in evaluation.