Your team is choosing how to add a capability to a Claude application. You want to apply the appropriate option, whether built-in tool, custom tool, Skill, or MCP server, based on the use case.
You would choose the option that...

• A. Is the most familiar to the team based on prior experience with similar capabilities.
• B. Is the newest available option among built-in tools, custom tools, Skills, and MCP servers.
• C. Matches the use case's scope, reuse needs, and integration context.
• D. Is the easiest to build given the team's current development tools and workflows.

The Claude application your team built has grown over six months, and the prompt-handling code has accumulated duplication and tangled control flow. The functionality is working, but new features are getting harder to add.
How would you address this?

• A. Plan a refactoring pass to consolidate duplicated logic, separate concerns, and simplify control flow before adding new features.
• B. Move all the prompt-handling code into a single large function to reduce the number of files developers have to navigate when reading the code.
• C. Refactor the prompt-handling code in small increments as part of each new feature ticket, treating the cleanup as a side effect of feature work.
• D. Continue adding features and plan a refactoring pass after the next two release cycles when the team has more bandwidth for internal work.

Your Claude application requests structured JSON output from the model. Most of the time the JSON is well- formed, but occasionally Claude returns malformed JSON that breaks downstream processing.
How would you handle the malformed output?

• A. Retry the same request repeatedly until valid JSON appears in the model's response, with the retry loop adding delay to the application's response time on affected requests.
• B. Switch to free-form text output so the application no longer depends on JSON parsing for any of the responses it sends to downstream systems during normal operation.
• C. Add output validation that parses Claude's response against the expected schema and treats malformed output as a recognized error path with retry or fallback handling.
• D. Manually inspect every response before downstream processing so a human reviewer catches any malformed JSON before the application passes the response to downstream systems.

You are designing a Claude application that processes user-submitted text. Some of that text could include sensitive information such as account numbers or passwords that the application should not send to Claude.
How would you design the application?

• A. Apply filtering at the boundary for the most commonly observed sensitive data patterns and expand coverage to additional patterns based on findings from production monitoring.
• B. Add a prompt instruction in the system prompt specifying the categories of sensitive information Claude should disregard when processing user-submitted text.
• C. Define the application boundary explicitly, identify what content can leave the boundary for Claude, and add filtering or redaction at the boundary.
• D. Log all user-submitted text before it is sent to Claude and review the logs periodically to identify whether sensitive information is reaching the model.