1. You are here:
2. Handbook
3. Security at GitLab
4. Security Assurance
5. Field Security Team
6. Independent Security Assurance

On this page

• Overview
  • Third Party Security Ratings
  • Third Party Vulnerability and Penetration Tests
    • SaaS
    • Self-Managed

Overview

GitLab contracts with third parties to conduct annual network and application penetration testing and perform continuous public security scanning (BitSight). Evidence of these are shared via our Customer Assurance Package.

Third Party Security Ratings

BitSight is a third party security rating platform that utilizes public information collected across multiple domains to provide a numeric score from 250-900 (similar to a credit rating, but security focused). GitLab publishes three reports using BitSight:

• GitLab Production- this report represents the security posture of any URL or IP associated with supporting our production architecture. Our goal is to maintain a rating of 700 or higher. Alerts from BitSight are monitored in real time and addressed as part of our Vulnerability Management program.
• GitLab Sub-Production- this report represents the security posture of any URL or IP that does NOT support the Production Environment noted above. This generally includes test environments and/or environments that are purposely kept out-of-date for quality assurance activities. This score is not monitored by GitLab and does not represent GitLab's Security Posture.
• GitLab User-Managed- many of our customers utilize static pages as part of gitlab.io. These pages are managed solely by our customers. However, from time to time, these pages are associated with GitLab erroneously. As such, we have moved any page on the GitLab.io domain to this report. This score is not monitored by GitLab and does not represent GitLab's Security Posture.

A copy of the Quarterly BitSight report and other Security Assurance artifacts are available to download in our Customer Assurance Package.

Third Party Vulnerability and Penetration Tests

SaaS

Customers are NOT authorized to conduct Vulnerability Scans or Penetration tests on GitLab's SaaS Application. A penetration test determines whether or not defensive measures employed are strong enough to prevent security breaches. If our Customers or Prospects were to attempt their own penetration testing of our SaaS environment, it could appear as a real incident to GitLab. For more information, please review GitLab's Terms of Service.

GitLab conducts external, independent penetration tests of our production architecture at least annually. In lieu of this, a report of our Annual Penetration Test is available in our Customer Assurance Package.

GitLab maintains a comprehensive Vulnerability Management program that is configured to identify vulnerabilities throughout our production architecture. Details are available within our Vulnerability Management handbook page.

Self-Managed

Penetration testing and Vulnerability scanning performed by self-managed customers must utilize the standard omnibus deployment to prevent false positives based on custom configurations. Automated vulnerability scanners commonly produce low priority issues and/or false positives. Before submitting the results from a scanner, please take a moment to confirm that the reported issues are actually valid and exploitable. Identified and validated vulnerabilities can be submitted through our HackerOne reporting program or by creating an issue for our security team.