최신Salesforce Certified Agentforce Specialist (AI-201) - Agentforce-Specialist무료샘플문제

A. Grounding data from external websites to augment a prompt with RAG.

B. Returning relevant data from the vector database to augment a prompt.

C. Modifying and updating data within the source systems connected to Data Cloud.

A. Use absolute keywords like "Always " and "Never" in Topic Instructions to enforce logic, such as
"Always escalate when severity is high" and "Never create a support case when severity is high".

B. Write the triage and routing logic in Topic Instructions using an IF, THEN, ELSE pattern: "Escalate to human service rep if the request is considered severe, otherwise create support case".

C. Create a custom variable severityLevel populated by a Triage action. Add filters so the "Escalate to human service rep" action only runs when severityLevel = 'High', and the "Create Support Case" action runs only when severityLevel != 'High'.

A. Ensure queries used for grounding employ offset so the token limits of models are not exceeded.

B. Review the token limit in the Einstein Trust Layer.

C. Review the model limitation in Prompt Builder versus the grounding data size.

A. Response times, accuracy and relevance of answers, and resolution success

B. Agent performance, token usage, and conversation duration

C. Response performance, tone, and CSATs

A. Activities such as tasks and events are excluded.

B. Empty data, such as fields without values or sections without limits, is filtered out.

C. Email addresses associated with the object are excluded.

A. Use the Agent API to start the downstream agent ' s session and pass the verified customer ID as a read- only context variable, ensuring security and preventing LLM alteration.

B. Store customer identity information in conversation variables created by the first agent and have other agents read those same conversation variables.

C. Implement a custom object to temporarily store verification status and have each agent query it via SOQL actions during execution.

A. Turn on Service Replies, Service AI Grounding, and Grounding with Knowledge.

B. Turn on Service AI Grounding, Grounding with Case, and Service Replies.

C. Turn on Service AI Grounding and Grounding with Knowledge.

A. View the user click path that led to each agent action.

B. Generate details reports on all agent conversations over any time period.

C. View session data including user input and agent responses for sessions.