In a Claude Code project, what is the primary purpose of a CLAUDE.md file at the repo root?
The .mdc Configuration File System
Loading lesson content...
Scenario: Claude Code for Continuous Integration You are integrating Claude Code into your CI/CD pipeline. The system runs automated code reviews, generates test cases, and provides feedback on pull requests. You need to design prompts that provide actionable feedback and minimize false positives. A new CI job asks Claude Code to improve coverage for a data normalization utility. The first generated tests pass, but they miss null inputs, timezone boundary cases, and duplicate IDs. Follow-up prompts like "be more thorough" produce inconsistent additions and occasional regressions. What workflow should you use to make improvement more reliable?