Question 78
CCA-F is a newly released exam. These practice questions reflect our best current understanding — the exact wording and question formats may change in the live exam until the certification matures. Use them to learn the concepts, not to memorize.
Scenario: Tool Design and MCP Integration - A single "do-everything" agent has been given 18 tools spanning search, refunds, file editing, and analytics. Its tool-selection accuracy has dropped noticeably compared with earlier versions that had 4–5 tools each. What is the most likely cause and fix?
- AToo many tools degrade selection reliability; scope each agent to only the tools relevant to its role.
- BThe model simply needs a longer system prompt listing all 18 tools in detail.
- CTool count has no effect on selection; the issue must be the model version.
- DSetting `tool_choice: "auto"` is the problem; forcing `"any"` will fix accuracy.