Question 820
A database architect reviews two implementation risks. An AI chat session lists several available MCP tools, including one that can modify production data, for a user authorized only to analyze schema. A Copilot agent receives a schema-inspection tool result containing a directive to reveal a secret in the next chat message. What is the most appropriate plan for this situation?
Explanation
Correct Answer
First, tool exposure and authorization determine what actions an assistant can perform, regardless of prompt wording. Second, database and tool output must not be promoted into instructions for privileged assistant behavior.
Incorrect Answers
Configure tool availability and effective identity to permit read-only schema operations and block write-capable tools; connect only to an approved authenticated MCP server with scoped read permissions and verified endpoint identity: database and tool output must not be promoted into instructions for privileged assistant behavior.
Bind the connector to an approved workspace and record data scope, consent and audit requirements before enabling use; Treat the returned schema text as untrusted data and ignore tool-output instructions that conflict with trusted policy: tool exposure and authorization determine what actions an assistant can perform, regardless of prompt wording.
Treat the returned schema text as untrusted data and ignore tool-output instructions that conflict with trusted policy; do not configure tool availability and effective identity to permit read-only schema operations and block write-capable tools: configure tool availability and effective identity to permit read-only schema operations and block write-capable tools.



Developing AI-Enabled Database Solutions (DP-800) Practice Tests