You are implementing a custom tool for your Claude agent. The tool needs to interact with an external pricing service that returns product data.
Which of the following best practices would you apply as you develop this tool?

• A. Define the tool with a loose schema and let the model interpret the inputs flexibly on each call the agent makes.
• B. Omit the tool description and let the model infer when to use the tool based on the tool's name and the rest of the prompt context.
• C. Define the tool with a clear schema, write a precise description for when to call it, and handle pricing service errors explicitly.
• D. Implement the tool with no error handling and let the agent loop catch failures whenever the pricing service returns an error during operation.

A team has deployed a multi-agent system in which a primary agent decomposes user requests and delegates subtasks to three specialized subagents: one for data retrieval, one for analysis, and one for report generation.
In production, the team observes that subagents are making redundant tool calls, occasionally exceeding token budgets, and sometimes producing outputs that contradict each other - all of which the primary agent passes along without catching.
What is the most appropriate way to address these failures?

• A. Give each subagent read access to the other subagents' outputs so they can identify and resolve contradictions without routing back through the primary agent.
• B. Add retry logic to each subagent so that when a tool call fails, the subagent retries automatically before escalating - and configure each subagent to log its tool calls and outputs to a shared trace so the team can audit redundancy and contradictions after the fact.
• C. Collapse the three subagents into a single large-context model call that handles retrieval, analysis, and generation in one pass.
• D. Strengthen the primary agent's management layer to enforce per-subagent tool budgets, validate outputs against a defined schema before passing them forward, and establish explicit handoff contracts between stages.

You are choosing between using STDIO-based communication and HTTP-based communication for an MCP server. The server will be invoked by a Claude Code session running locally.
Which communication pattern would you use?

• A. HTTP on the grounds that HTTP is preferable to STDIO in any context the application might encounter during normal operation.
• B. HTTP polling against the MCP server, treating HTTP as a substitute for the standard MCP communication patterns.
• C. STDIO because STDIO is well-suited to local process communication and avoids unnecessary network setup.
• D. STDIO and HTTP simultaneously for redundancy, with the team running two communication patterns at the same time across Claude Code sessions.

Your Claude application's token costs have grown faster than expected. The team has not been tracking token usage by feature, so the team cannot identify which features are driving cost. The team is debating how to respond.
How would you respond?

• A. Reduce token usage uniformly across all features by half, applying the cut evenly across the application during normal operation.
• B. Switch every feature to the smallest model to cut cost broadly across the application during normal operation.
• C. Tell the team that cost growth is unavoidable as the application scales and that no investigation will change the trajectory of the application's cost over the next several quarters.
• D. Add token usage tracking by feature to the application's logging so the team can identify which features drive cost before recommending changes.

You are building a Claude application that processes 10,000 customer emails overnight to extract structured data. The work is non-interactive, runs once daily, and has a flexible completion window of several hours.
Which Claude API would you use?

• A. The Batch API, which is designed for non-interactive workloads with flexible completion windows.
• B. The real-time Messages API, processing the emails one at a time sequentially to ensure consistent ordering of results.
• C. The real-time Messages API with concurrent requests to process the emails as fast as possible during the overnight window.
• D. The streaming responses API to process each email and return partial results to a database as the model generates them.