CCA-F: Claude Certified Architect — Foundations

By Anthropic · Question Mode

QuestionQ119

Context Management & Reliability

You are incorporating Claude Code into your Continuous Integration/Continuous Deployment (CI/CD) pipeline. The system performs automated code reviews, creates test cases, and gives feedback on pull requests. You must design prompts that deliver actionable feedback while minimizing false positives.

A developer uses Claude Code to refactor a function during a development session. Before committing, they ask that same Claude session to review the code for issues. Later, an independent automated CI review finds several bugs that the same-session review missed.

What most effectively explains this discrepancy?

Explanation

A review performed in the same session carries forward the implementation context and earlier reasoning. That prior context can anchor the model to its own choices, making it less likely to independently challenge those decisions and identify defects. A separate CI review starts from an independent context, which can expose bugs missed during self-review. Claude Code supports resuming sessions specifically to retain conversational context.

Learn more

Community Discussion

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