Rubric
Contents — domains, guide and mocks

Claude Code Configuration & Workflows

CCAR-F · Domain 324 questions · 20% of the exam

Answer everything, then check. Each result links back to the lesson for the objective it came from.

0 of 24 answered0:00
  1. Question 1 · 3.1

    A platform team’s Claude Code sessions follow the house error-handling convention, but the same repository’s nightly CI job, which runs Claude Code headlessly, ignores it. The convention was written by the team lead months ago.

    What is the most likely cause?

  2. Question 2 · 3.1

    A healthcare startup’s root CLAUDE.md has grown to 700 lines: general rules, 200 lines of React guidance, 150 lines on database migrations and a long release procedure. Engineers report that Claude often ignores rules near the bottom.

    Which two changes will most reduce what loads into every session while keeping the guidance available? (Select 2.)

  3. Question 3 · 3.1

    Claude Code is launched in repo/services/billing/. Which CLAUDE.md files are in context at launch?

  4. Question 4 · 3.1

    An insurer’s security team requires that Claude Code never reads files in secrets/ on any developer machine. A developer proposes adding “Never open anything in secrets/” to the managed-policy CLAUDE.md.

    What is the best response?

  5. Question 5 · 3.2

    A payments team creates .claude/skills/release/SKILL.md, which tags a version and publishes it to the package registry. A week later, during an unrelated refactor, Claude runs the release skill on its own after tests pass.

    Which change prevents this while keeping /release available to engineers?

  6. Question 6 · 3.2

    A developer writes a skill with allowed-tools: Read Grep Glob, expecting it to be read-only. During a run, Claude edits a file after the user approves an Edit permission prompt.

    What explains this behaviour?

  7. Question 7 · 3.2

    A hospital software team has a /audit-phi-logging skill that scans hundreds of service files for places where patient identifiers might be logged. After it runs, developers find their session slow and cluttered.

    What is the most appropriate change?

  8. Question 8 · 3.2

    A repository has both .claude/commands/lint.md and .claude/skills/lint/SKILL.md. What happens when an engineer types /lint?

  9. Question 9 · 3.3

    A logistics company keeps React components in web/src/components/ and its GraphQL resolvers in server/src/resolvers/. Each has its own detailed conventions, currently all in a 450-line root CLAUDE.md.

    Which approach best loads each set of conventions only when relevant?

  10. Question 10 · 3.3

    A fintech team writes a rule with paths: ["src/payments/*.ts"]. Claude follows it when editing src/payments/refund.ts, but ignores it when working on src/payments/providers/stripe.ts.

    What is the fix?

  11. Question 11 · 3.3

    A university’s research-software team wants conventions for database migration files, which live in db/migrations/ in six different services across the repository.

    Which two statements are correct? (Select 2.)

  12. Question 12 · 3.3

    A rule file in .claude/rules/style.md has no frontmatter at all. When does it load?

  13. Question 13 · 3.4

    A product team needs to move session storage from in-memory to Redis. The change touches the auth service, two API gateways and the test harness, and there is debate about whether to use a cache-aside or write-through design.

    How should the engineer approach this with Claude Code?

  14. Question 14 · 3.4

    A stack trace points to a null dereference on line 88 of invoice_pdf.py. The fix is a guard clause, and the file has a unit test suite.

    What is the most efficient approach?

  15. Question 15 · 3.4

    During plan mode, Claude proposes a plan to split a monolith’s reporting module into a separate service. The engineer agrees with most of it but wants the database migration moved to the end and one endpoint left untouched.

    What should the engineer do?

  16. Question 16 · 3.4

    Which statement about plan mode is accurate?

  17. Question 17 · 3.5

    An insurance team asks Claude to “format policy numbers consistently”. Across three attempts Claude produces three different formats, each defensible. The team lead’s fourth message is “No — CONSISTENTLY, like our real policy numbers!”

    What is the most effective next step?

  18. Question 18 · 3.5

    A developer has corrected Claude four times in one session about how a caching layer should invalidate entries. Each fix introduces a new problem, and Claude now references approaches that were already rejected.

    What does the Claude Code guidance recommend?

  19. Question 19 · 3.5

    A fintech team wants Claude to implement a fee calculator with tiered rates, rounding rules and a cap. Previous attempts from a written description passed a quick manual check but failed in staging on edge cases.

    Which two changes will most improve the next iteration? (Select 2.)

  20. Question 20 · 3.5

    After a database column is renamed, three things break in a reporting service: a SQL query, the ORM model that maps it, and a CSV export that reads the model. A developer plans to report them to Claude one per message.

    What is the better approach, and why?

  21. Question 21 · 3.6

    A team adds claude "Review the changes in this branch" as a step in their CI pipeline. The job never finishes and is killed by the runner’s timeout.

    What is the most likely fix?

  22. Question 22 · 3.6

    A platform team wants review findings posted as inline comments on the exact lines of a pull request, and wants the build to fail on any high-severity finding.

    Which approach best supports this?

  23. Question 23 · 3.6

    A CI job asks Claude Code to generate unit tests for new code, then, in the same session, asks it to review those tests for gaps. Reviewers keep finding weak assertions the review step approved.

    What is the best improvement?

  24. Question 24 · 3.6

    A security-conscious fintech wants a nightly job that reads the codebase and reports dependency risks. It must never edit files or run arbitrary shell commands, and nobody will be watching.

    Which configuration fits best?

You can change answers until you check. Nothing is saved or sent anywhere.