Get real CCAR-F exam questions for better preparation

Anthropic CCAR-F Exam Overview:

Anthropic CCAR-F Exam Syllabus Topics:

Anthropic Claude Certified Architect - Foundations Sample Questions:

Question 1

You are building a structured data extraction system using Claude. The system extracts information from unstructured documents, validates the output using JavaScript Object Notation (JSON) schemas, and maintains high accuracy. It must handle edge cases gracefully and integrate with downstream systems.
Monitoring shows 12% of extractions fail Pydantic validation with specific errors like "expected float for quantity, got `2 to 3'". Retrying these requests without modification produces identical failures.
What's the most effective approach to recover from these validation failures?

A. Pre-process source documents to standardize problematic formats before sending them for extraction.
B. Set temperature to 0 to eliminate output variability and ensure consistent formatting.
C. Implement a secondary pipeline using a larger model tier to reprocess documents that fail validation.
D. Send a follow-up request including the validation error, asking the model to correct its output.

Question 2

You are building developer productivity tools using the Claude Agent SDK. The agent helps engineers explore unfamiliar codebases, understand legacy systems, generate boilerplate code, and automate repetitive tasks. It uses the built-in tools (Read, Write, Bash, Grep, Glob) and integrates with Model Context Protocol (MCP) servers.
You've configured your Claude agent with three MCP servers: one for git operations, one for Jira ticket management, and one for documentation search.
When a user asks the agent to "create a branch for JIRA-123 and add documentation links to the ticket," how does the agent access tools across these servers?

A. You must specify which MCP server to use for each turn, and the agent can only access one server's tools at a time.
B. The agent queries each server sequentially to determine which handles each tool, routing calls based on tool name prefixes.
C. Tools from all configured MCP servers are discovered at connection time and available simultaneously to the agent.
D. The agent automatically selects the most relevant server based on the request and loads only that server's tools.

Question 3

Your pipeline reviews every pull request using a single API call with a static prompt containing the diff and the full text of each changed file; unchanged files are not included. Reviews are posted asynchronously and do not block pull-request creation. Developers report that reviews consistently miss bugs involving cross-file interactions--for example, a pull request renames a function's parameters, but the review does not flag callers in other files that still use the old parameter names. Post-release analysis shows that cross-file bugs account for 35% of production incidents from reviewed pull requests. What is the most effective change to your review design?

A. Add chain-of-thought instructions asking the model to list all external references in the diff and then reason step by step about how each change might affect callers in other files.
B. Use static analysis to build a dependency graph of changed code, and then expand the prompt to include every file within two dependency hops of any changed file.
C. Redesign the review as a turn-limited agentic task in which the model can read files and search the codebase through tools, following references to verify cross-file findings.
D. Run parallel review passes for each changed file with its direct dependents included, and then aggregate and deduplicate the findings through a final summarization call.

Question 4

Which practice MOST improves prompt maintainability?

A. Minimal punctuation.
B. Random ordering.
C. One paragraph containing all instructions.
D. Clearly separated sections with headings.

Question 5

Your MCP server includes archive_file(file_id) and delete_file(file_id) tools. Production logs show the agent calls delete_file when users ask to "remove old backups," but company policy requires archiving backup files. Both tools currently have minimal descriptions: "Archives a file" and
"Deletes a file." Which change most directly improves tool selection for this scenario?

A. Implement server-side validation that rejects delete_file calls for files tagged as backups, returning an error message suggesting archive_file.
B. Add a confirmation step that requires users to type "CONFIRM DELETE" before delete_file executes.
C. Add few-shot examples to the system prompt demonstrating that requests involving "backup" or
"old" should use archive_file.
D. Expand tool descriptions to clarify use cases, adding guidance like "Do not use for backup files" to delete_file.

Solutions: