Summary

Completed

In this module, you:

• Built a working definition of agentic AI in the SDLC and learned how agents differ from assistants.

• Learned how agents show up in GitHub as contributors through branches, pull requests, workflow runs, and reviews.

• Practiced the plan → act → evaluate lifecycle as the core model for agent execution and iteration.

• Learned how GitHub serves as a system of record and a control plane, using controls like rulesets/branch protection, required checks, required reviews, CODEOWNERS, and environments (when configured).

• Identified common risks and anti-patterns, and learned how traceability plus a contributor-based review model helps you evaluate agent work reliably.

Learn more

For deeper reading, use official GitHub documentation on:

• Reviewing a pull request created by GitHub Copilot

• Creating rulesets for a repository and About protected branches (branch protection rules)

• About code owners (CODEOWNERS)

• Use GITHUB_TOKEN for authentication in workflows

• Uploading a SARIF file to GitHub (code scanning)

• About push protection (secret scanning)

• Audit log for an enterprise (availability depends on organization/enterprise configuration)