2025 年 12 月，我按照 Agentforce Specialist 的考试大纲（Exam Guide），用 NotebookLM 基于 Trailhead 资料生成了 4 期考点解析音频，发在公众号「SalesforceTalk」上，每期 17 到 18 分钟。这里把每期对应的考点和权重整理到一起，音频点链接到公众号收听。
1. AI Agents（35%）
收听：Agentforce Specialist - AI Agents 考点解析（17:25）
- Given a use case, manage deterministic behavior for the agent using filters and variables.
- Explain how an agent works and how the reasoning engine powers Agentforce.
- Given a scenario, select and configure standard topics, custom topics, standard agent actions, and custom agent actions based on agent types.
- Explain how to manage Agentforce security, including the concept of the Agent User and how it applies to an Employee Agent, Service Agent, or Sales Agent.
- Given a scenario, identify when to use an Employee agent, Service agent, or Sales agent.
- Explain the process for connecting agents to various channels such as digital experience, email, and Slack.
2. Prompt Engineering（20%）
收听：Agentforce Specialist - Prompt Engineering 考点解析（18:09）
- Given business requirements, identify when it’s appropriate to use Prompt Builder.
- Identify the right user roles to manage and execute prompt templates.
- Identify the considerations for creating a prompt template using field generation and flex types.
- Given a scenario, identify the appropriate grounding technique.
- Explain the process for creating, activating, and executing prompt templates.
- Explain how to use best practices for writing effective prompts.
3. Data Cloud for Agentforce（20%）
收听：Agentforce Specialist - Data Cloud for Agentforce 考点解析（17:24）
- Explain the considerations of Agentforce Data Library and its types.
- Given a scenario, improve an agent’s response with unstructured data using chunking and indexing.
- Identify the considerations for retrievers in Data Cloud such as individual and ensemble.
- Given a scenario, identify the considerations for search type such as keyword, vector, and hybrid.
4. Development Lifecycle（20%）和 Multi-Agent Interoperability（5%）
收听：Agentforce Specialist - Development Lifecycle & Multi-Agent 考点解析（17:49）
Development Lifecycle（20%）
- Given a scenario, test an agent using Agentforce Testing Center.
- Identify the considerations for deploying an agent from sandbox to production.
- Explain the process for managing and monitoring agent adoption.
Multi-Agent Interoperability（5%）
- Explain the purpose of Model Context Protocol (MCP) and its use cases.
- Explain the purpose of agent to agent protocol. Given a scenario, identify when it’s appropriate to use Agent API.
旧版：AI Specialist（2024）
Agentforce Specialist 的前身是 AI Specialist。2024 年 10 月做过两期，同样由 NotebookLM 生成，每期约 9 分钟。部分内容已经过时，仅供参考：
- AI Specialist Certification - Einstein Trust Layer（8:50）
- AI Specialist Certification - Use Generative AI in CRM（8:48）
音频首发于微信公众号「SalesforceTalk」。