本橋孝昭です
今回は2026年5月22日（Spring26）ごろにAgenetforceスペシャリストの試験範囲が変更されたので解説します
※試験内容変更日は2026年6月時点ではすでにページでは表示されてないですが、以下のような2026年5月22日に変更された記述は確認できました
変更内容
変更された内容についてみてきましょう
試験概要
Here are details about the Salesforce Certified Agentforce Specialist Exam.
Content: 60 multiple-choice questions and up to five unscored questions
Time allotted to complete the exam: 105 minutes
Passing score: 72%
Version: Exam questions align to the Spring ’26 release.
Registration fee: USD 200, JPY 30,000, plus applicable taxes as required per local law; JPY 30,000
Retake fee: USD 100, JPY 30,000, plus applicable taxes as required per local law; JPY 30,000
Delivery options: Proctored exam delivered onsite at a testing center or in an online proctored environment; find more information on scheduling an exam here.
References: No hard-copy or online materials may be referenced during the exam.
Prerequisite: None
Language offerings: English, French, Japanese, and Spanish
This exam may contain up to five additional unscored questions to gather performance data. Unscored questions are randomly integrated and have no impact on your final exam result.
試験範囲
Exam Outline
The Salesforce Certified Agentforce Specialist Exam measures a candidate’s knowledge and skills related to the following objectives.
Prompt Engineering: 20%
Given business requirements, identify when it's appropriate to use Prompt Builder.
Identify access controls governing prompt templates.
Identify the considerations for using a prompt template type such as 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 implement best practices for writing effective prompts.
Identify the security and privacy features of the Trust Layer.
Explain how to manage and prevent specific models from being accessed.
Data 360 Fundamentals: 20%
Explain the considerations of Agentforce Data Library and its concepts.
Explain foundational concepts of Data 360 such as chunking, indexing, and retrievers.
AI Agents: 35%
Explain how an agent works and its basic building blocks of agent script.
Explain the components and benefits of hybrid reasoning, including how Agent Script functions in Canvas and Script View.
Given a use case, manage deterministic behavior for the agent using mechanisms like filters, variables, and template expressions.
Given a scenario, select and configure standard topics, custom topics, standard Agent actions, and custom Agent actions.
Explain the process for connecting agents to various channels such as digital experience, email, voice, and Slack.
Explain the security context in which the agent is actually running, and how it impacts agent action execution.
Given a scenario, identify when to use an Employee or Service agent.
Given a scenario, identify when it's appropriate to use Agent API.
Testing, Deployment, & Maintenance: 10%
Given a scenario, test an agent using Testing Center.
Explain how Testing Center evaluations work.
Identify the considerations for deploying an agent from sandbox to production.
Identify the considerations for deploying a template from sandbox to production.
Governance & Observability: 10%
Explain the process for managing and monitoring agents.
Explain agent analytics and agent optimization
Multi-Agent Orchestration: 5%
Given a scenario, determine whether a Single-Agent (SOMA) architecture is appropriate for scalability and control.
Explain the purpose of existing open standard multi-agent protocols such as MCP and A2A.
過去の試験内容はこちらからご確認ください