Practice Free CCA-F Exam Online Questions
You are analyzing a massive legacy codebase and want to explore an alternative, potentially destructive refactoring approach without polluting your main session’s context window.
Which session management feature should you use?
- A . – resume
- B . fork session
- C . /compact
- D . /clear
When a tool lookup (e.g., get_customer) returns multiple matches, what is the most reliable way for the agent to proceed?
- A . Automatically escalate to a human agent.
- B . Select the most recently active account using a heuristic.
- C . Instruct the agent to ask the user for additional identifiers for clarification.
- D . Return a structured error and terminate the workflow.
Which of the following scenarios represent architecturally appropriate use cases for implementing hooks in a production Claude application? Choose 2 correct answers.
- A . Normalizing tool outputs, such as date formats or status codes, before they are appended to the context window.
- B . Guaranteeing that an extracted JSON object has no semantic reasoning errors.
- C . Ensuring Claude maintains a conversational and friendly tone with customers.
- D . Intercepting tool calls to deterministically block policy-violating actions before they execute.
When using tool_use with a strict JSON schema for data extraction, what does the Claude API explicitly guarantee?
- A . Semantic correctness of the extracted data.
- B . Structural compliance, effectively eliminating JSON syntax errors.
- C . 100% factual accuracy of the extracted entities.
- D . Elimination of the need for downstream data validation.
When designing subagents to handle conflicting data from different sources during the final synthesis phase, which of the following are recommended practices? Choose 2 correct answers.
- A . Silently average the conflicting numerical values to provide a balanced estimate.
- B . Require subagents to output structured claim-source mappings.
- C . Preserve information provenance including the source, confidence level, and timestamp.
- D . Arbitrarily select the value from the most recently updated database to resolve the conflict.
To integrate an MCP server providing database tools for a team, you must configure it so that all developers share the integration but do not leak secrets into version control.
What is the correct configuration approach?
- A . Configure the server in the user-scoped ~/. claude. json file to keep credentials local.
- B . Configure the server in the project-scoped .mcp. j son file using environment variable expansion like ${DB_TOKEN}.
- C . Hardcode an encrypted API token directly in the .mcp.json file.
- D . Store the credentials in an ignored secrets. json file referenced by CLAUDE. md.
When passing context between agents in a multi-agent orchestration, why is it highly recommended to use structured data formats?
- A . To automatically trigger JSON schema validation by the Claude API.
- B . To separate content from metadata (such as source URLs, document names, and page numbers) to preserve proper attribution.
- C . To compress the token count of the prompt by 50% using the Message Batches API.
- D . To bypass the maximum token limit of the subagent’s isolated context window.
What is the recommended architectural pattern to prevent the loss of critical transactional details (like account IDs, exact amounts, and dates) during long-running interactions?
- A . Use the /compact command automatically after every 10 turns.
- B . Extract these facts into a persistent, immutable ‘case facts’ block positioned at the start of the context.
- C . Progressively summarize the conversation history and rely on the model’s extended thinking parameters.
- D . Switch to a higher-tier model with a 1 million token context window.
After executing a requested tool in an agentic loop, how must the application handle the follow-up request to ensure the model can accurately reason about the next action?
- A . Pass only the new tool result to save token space in the context window.
- B . Pass the tool result inside an assistant message equipped with an is_error flag.
- C . Append a user message containing a tool_result block with the matching tool_use_ld, and include the full prior conversation history.
- D . Use the /compact command to summarize the tool output before returning it to the loop.
To optimize a human-in-the-loop workflow with limited reviewer capacity, how should you determine which documents require manual review?
- A . Rely on the model’s self-reported conversational confidence score in the output.
- B . Have models output field-level confidence scores and calibrate review thresholds using labeled validation sets.
- C . Escalate any document that exceeds 5 pages in length.
- D . Route documents randomly based on a flat 10% sample rate.