A teammate is debugging a Claude application whose system prompt has grown to several hundred lines and now contains overlapping, contradictory, and obsolete instructions.
How would you advise the teammate?

• A. Tighten only the contradictory rules first, treating the overlap and obsolete content as lower-priority work the team can address later.
• B. Add more explicit instructions so the most recent rules dominate the model's interpretation of the prompt during each request.
• C. Audit the prompt for overlap, contradiction, and obsolete content, then refactor so each instruction is clear, current, and non-redundant.
• D. Split the prompt across multiple system prompts so the model sees a smaller portion at any given time.

A teammate has asked you to explain why the team's Claude application is billed for output tokens at a different rate than input tokens. They had assumed the rate was the same for both.
How would you explain the difference?

• A. Output tokens are not billed at all, because cost is determined entirely by the input tokens sent to the model on each request.
• B. Output tokens are typically billed at the same rate as input tokens, and the apparent rate difference is a billing error to report to Anthropic.
• C. Output tokens are typically billed at a higher rate than input tokens, and cost models for the application should reflect both rates separately.
• D. Output tokens are typically billed at a lower rate than input tokens, because output tokens are cheaper to produce than input tokens are to process.

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

A new agent your team built handles customer support tickets, but it routinely gets confused when a single ticket spans billing, shipping, and product issues. The agent often loses track of which sub-issue it has already addressed and revisits the same one. The team is considering architectural changes.
What architectural change would you recommend?

• A. Increase the size of the agent's context window so it can hold the full ticket history at once.
• B. Add detailed prompting that instructs the agent to track which sub-issues have been resolved and which remain.
• C. Switch to a deterministic workflow that handles billing, shipping, and product issues in a fixed sequence.
• D. Introduce an orchestrator agent that delegates billing, shipping, and product sub-issues to dedicated subagents.

Your Claude application returns confident-sounding answers, but occasionally those answers contain factual errors that downstream systems treat as ground truth. The team is concerned about the application's confidence-versus-accuracy gap.
How would you address the gap?

• A. Lower the model's temperature so the model's responses sound less confident and downstream systems are less likely to treat the responses as ground truth in normal operation.
• B. Reject every response the application produces until a manual accuracy review is conducted on each response by a human reviewer before any downstream system uses it.
• C. Add a disclaimer to every output telling users to verify the accuracy of the output and treat the disclaimer as the primary mechanism for managing the confidence-versus-accuracy gap.
• D. Apply skepticism toward confident output by adding validation steps, sourcing requirements, or confidence calibration before treating outputs as ground truth.