This article is part of a series on adopting GitHub Advanced Security at scale. For the previous article in this series, see "Phase 2: Preparing to enable at scale."
We recommend you identify a few high-impact projects or teams to use in a pilot rollout of GHAS. This allows an initial group within your company to get familiar with GHAS and builds a solid foundation for GHAS before you roll it out to the remainder of your company.
The steps in this phase will help you enable GHAS on your enterprise, begin using its features, and review your results. If you’re working with GitHub Expert Services, they can provide additional assistance through this process through onboarding sessions, GHAS workshops, and troubleshooting as needed.
Before you start your pilot projects, we recommend that you schedule some meetings for your teams, such as an initial meeting, midpoint review, and a wrap-up session when the pilot is complete. These meetings will help you all make adjustments as needed and ensure your teams are prepared and supported to complete the pilot successfully.
You need to enable GHAS for each pilot project, either by enabling the GHAS features for each repository or for all repositories in any organizations taking part in the pilot. For more information, see "Managing security and analysis settings for your repository" or "Managing security and analysis settings for your organization"
You can quickly configure default setup for code scanning across multiple repositories in an organization using security overview. For more information, see "Configuring default setup for code scanning at scale."
You can also choose to enable code scanning for all repositories in an organization, but we recommend configuring code scanning on a subset of high-impact repositories for your pilot program.
For some languages or build systems, you may need to instead configure advanced setup for code scanning to get full coverage of your codebase. However, advanced setup requires significantly more effort to configure, customize, and maintain, so we recommend enabling default setup first.
If your company wants to use other third-party code analysis tools with GitHub code scanning, you can use actions to run those tools within GitHub. Alternatively, you can upload results, which are generated by third-party tools as SARIF files, to code scanning. For more information, see "Integrating with code scanning."
GitHub scans repositories for known types of secrets, to prevent fraudulent use of secrets that were committed accidentally.
You need to enable secret scanning for each pilot project, either by enabling the feature for each repository or for all repositories in any organizations taking part in the project. For more information, see "Managing security and analysis settings for your repository" or "Managing security and analysis settings for your organization."
Next, enable push protection for each pilot project.
If you plan to configure a link to a resource in the message that's displayed when a developer attempts to push a blocked secret, now would be a good time to test and start to refine the guidance that you plan to make available.
Start to review activity using the push protection metrics page in security overview. For more information, see "Viewing metrics for secret scanning push protection in your organization."
If you have collated any custom patterns specific to your enterprise, especially any related to the projects piloting secret scanning, you can configure those. For more information, see "Defining custom patterns for secret scanning."
To learn how to view and close alerts for secrets checked into your repository, see "Managing alerts from secret scanning."
For the next article in this series, see "Phase 4: Create internal documentation."