CCAR-F scenarios › Claude Code for Continuous Integration · Question 44 of 60
Scenario 5 · Claude Code for Continuous Integration
Claude Code runs inside your CI/CD pipeline. It reviews pull requests automatically, proposes test cases, and posts feedback on the PR. The central challenge is writing prompts and configuration that produce actionable findings while keeping false positives low enough that developers keep trusting the output.
Claude Code Configuration & Workflows
Tests generated in CI are often low value, such as getter checks and assertions that a mock was called, and they frequently import fixture helpers that don't exist in the repository. The team has written guidance on what makes a test valuable and keeps a set of approved fixtures. Where should that guidance go so CI-invoked Claude Code uses it?
Show answer and explanation
Answer: C — In the project CLAUDE.md in the repository, documenting the testing standards, what counts as a valuable test, and the available fixtures.
CLAUDE.md is how CI-invoked Claude Code receives project context such as testing standards and fixture conventions, and keeping it in the repository keeps it versioned along with the code. (a) puts project knowledge in user-level config on one machine, outside version control, where it drifts from the repository. (b) controls the shape of the output, not the quality of the tests, and Claude would still be guessing about fixtures. (d) uses vague wording ("meaningful") without listing what the fixtures actually are.
More from this scenario
- Your pipeline has Claude Code generate a fix for a failing build and then, in the same…
- Select two. Developers say the automated PR review flags too many non-issues, such as…
- Your review flags issues in four categories: security, correctness bugs, performance…
- The same kind of problem, an unparameterized SQL string built from request input, is…
- Your pipeline posts a readable markdown summary of review findings on each PR. The…