CCA-F: Claude Certified Architect — Foundations

By Anthropic · Question Mode

QuestionQ123

Prompt Engineering & Structured Output

You are integrating Claude Code into your Continuous Integration/Continuous Deployment (CI/CD) pipeline. The system performs automated code reviews, generates test cases, and gives feedback on pull requests. You need to design prompts that deliver actionable feedback and reduce false positives.

Your test-generation process creates unit tests for new code, but reviews show that 55% are low-value: trivial assertions that merely confirm functions do not throw exceptions, tests that duplicate existing coverage, or tests that disregard your team’s fixture conventions.

How can you reduce the rate of low-value tests being generated in the first place?

Explanation

A repository CLAUDE.md supplies shared, project-specific instructions that Claude Code loads as context. Documenting concrete test-quality criteria, approved fixtures and their intended uses, and examples of meaningful behavioral tests gives the model the guidance needed to avoid trivial, duplicative, or convention-breaking tests during generation. Anthropic recommends using CLAUDE.md for team-shared project instructions and being specific about conventions and workflows.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!