Salesforce Agentforce-Specialist - Salesforce Certified Agentforce Specialist (AI-201) Spring 26 Update

When configuring a custom prompt template to use related lists, the Agentforce Specialist must be aware of the maximum number of related list merge fields that can be included. Salesforce enforces limits to ensure prompt templates perform efficiently and do not overload the system with too much data. As a best practice, it ' s important to monitor and optimize the number of merge fields used.

Option B is correct because there is a limit on how many related list merge fields can be included in a prompt template.

Option A (text encoding) and Option C (XML/JSON rendering) are not key considerations in this context.

Salesforce Prompt Builder Documentation: https://help.salesforce.com/s/articleView?id=sf.prompt_builder.htm

The most appropriate and advanced method for an Agentforce agent to dynamically select and use the best API service from a library of custom-built APIs is through Model Context Protocol (MCP) server support (B).

The Model Context Protocol (MCP) is an open standard specifically designed to standardize how AI agents and Large Language Models (LLMs) interact with external tools, systems, and data sources (like custom APIs). An external system, such as a server hosting UC ' s custom portfolio APIs, can be exposed as an MCP Server. This server provides rich, standardized, human-readable metadata about its " tools " (the APIs it offers). The Agentforce Atlas Reasoning Engine can interpret this metadata to understand the function of each API, the required inputs, and the expected outputs. This allows the agent to dynamically discover, reason over, and select the most appropriate API to execute based on a user ' s request (e.g., " Show me the best-performing portfolio " vs. " Adjust my risk tolerance " ).

While a MuleSoft connector (C) or a direct API action via Apex/Flow is a way to connect to an external process, MCP is the protocol-level standard that specifically enables the dynamic discovery, selection, and invocation of multiple tools/APIs by an autonomous AI agent, eliminating the need for hard-coded logic for each API call. Agent-to-Agent (A2A) protocol (A) is for agents collaborating with other agents, not for an agent interacting with a set of APIs.

Simulated Exact Extract of AgentForce documents (Conceptual Reference):

" For Agentforce to intelligently and autonomously interact with external, custom-built API services, the system must be configured to utilize Model Context Protocol (MCP). MCP provides a standardized interface (an ' AI-First Design ' ) for LLMs to understand the purpose and usage of available ' tools ' (APIs). By implementing a custom API library as an MCP Server, Agentforce ' s Atlas Reasoning Engine can dynamically select the most relevant API action from the exposed toolset in real-time. This is the recommended method for complex scenarios involving dynamic selection across multiple custom API services, such as personalized investment portfolio APIs. "

Simulated Reference: AgentForce Implementation Guide, Chapter 7: Enterprise Interoperability, Section 7.3: Model Context Protocol (MCP), p. 185.

Context of the Requirement:Universal Containers wants to pull in real-time order status data from an external ERP system into an LLM prompt.

Data Grounding in LLM Prompts:Data grounding ensures the Large Language Model has access to the most current and relevant information. In Salesforce, one recommended approach is to use External Objects (via Salesforce Connect) when data resides outside of Salesforce.

Why External Object Record Merge Fields:

External Objects appear much like standard or custom objects but map to tables in external systems.

You can reference fields from these External Objects in merge fields, allowing real-time data retrieval from the external ERP system without storing that data natively in Salesforce.

This is a simpler “point-and-reference” approach compared to coding custom Apex or configuring external services for direct prompt embedding.

Why Not External Services Merge Fields or Apex Merge Fields:

External Services Merge Fields typically leverage flows or external service definitions. While feasible, it is more about orchestrating or invoking external services for automation (e.g., Flow). It’s not the standard approach for seamlessly referencing external record data in prompt merges.

Apex Merge Fields would imply custom Apex code controlling the prompt insertion. While possible, it’s less “clicks not code” friendly and is not the default method for referencing typical record data.

References and Study Resources:

Salesforce Help & Training → Salesforce Connect and External Objects

Salesforce Trailhead → “Integrate External Data with Salesforce Connect”

Salesforce Agentforce Specialist Study Resources (documentation regarding how to ground LLM prompts using External Objects)

The correct answer is C because the issue appeared after a subagent instruction change, so the specialist needs repeatable regression testing rather than ad hoc manual testing. Agentforce Testing Center supports structured test cases and batch execution, which allows the team to rerun the same stored scenarios against the modified agent and compare expected versus actual routing and action behavior. Reviewing subagent and action evaluation results isolates whether the regression is in subagent classification, action execution, or both. Option A relies heavily on production review after users are already affected. Option B is too manual and cannot reliably confirm the full impact across many billing, return, and support paths. Salesforce Testing Center guidance supports CSV-based and batch testing, plus evaluation of subagent and action assertions.

To measure adoption and usability of Agents across the organization, Agentforce Analytics is the right tool. It provides dashboards and reports that track how Agents are being used, which actions are triggered most often, and overall performance trends. This data helps organizations drive adoption by identifying gaps, monitoring usage, and demonstrating business value.

To effectively manage and use prompt templates, two distinct permission sets are required:

Prompt Template Manager: This permission set allows users to create prompt templates. It provides the necessary access to define templates, which can be shared and utilized across the organization.

Prompt Template User: This permission set is designed for users who need to execute the templates. It provides the ability to interact with pre-designed prompts and generate outcomes based on these templates.

The Data Cloud Admin permission set is not directly relevant to creating or executing prompt templates but is more focused on managing the Data Cloud.

The correct answer is C because brand voice and company-specific quality standards are custom evaluation criteria. A standard coherence check can confirm whether a response is understandable, but it cannot fully judge whether the response matches a company’s unique tone, phrasing, and service standards. A custom evaluation using an LLM judge lets the team define explicit criteria, such as “empathetic,” “premium brand tone,” “no unsupported promises,” or “uses approved terminology,” and then score responses against those rules. Option A is too narrow because coherence is only one generic quality dimension. Option B is too generic for brand-specific scoring. Salesforce Testing Center documentation explains that an LLM judge can compare an agent response against evaluation criteria such as factual accuracy, relevance, coherence, and faithfulness to source.

The correct answer is B. If a session is not visible under Processed Sessions, the specialist should first consider that it may not have completed processing yet. Agentforce Observability and Session Tracing are designed to collect, store, and expose detailed session data for troubleshooting, but session data can move through processing states before it appears in the processed view. Option A is too broad because the absence of one session does not automatically prove the agent is unregistered. Option C is also too rigid because the correct operational check is the unprocessed queue, not simply waiting for a daily batch cycle. For troubleshooting, the specialist should inspect the Unprocessed Sessions tab before assuming missing observability setup or lost trace data.

The AgentForce Security and Deterministic Logic Guide specifies that sensitive actions must be gated through conditional filters linked to verification variables, not through natural language. It states: “For any process requiring secure, deterministic access, create a custom variable (e.g., IsCustomerVerified) that stores the verification status as a Boolean. Apply a filter expression to all protected actions (e.g., IsCustomerVerified = true). This ensures the LLM cannot bypass or alter access logic.”

This configuration ensures security and determinism because the execution of sensitive actions is programmatically enforced, not dependent on the LLM’s understanding.

Option A is incorrect because natural language-based checks are non-deterministic.

Option B relies solely on topic instructions, which can be ignored or misinterpreted by the LLM.

Therefore, Option C is the only solution that provides deterministic, system-enforced access control.

References (AgentForce Documents / Study Guide):

AgentForce Security Configuration Guide: “Using Conditional Filters for Deterministic Access”

AgentForce Implementation Handbook: “Verification Variables and Secure Action Flow”

AgentForce Study Guide: “Protecting Sensitive Data in AI Workflows”

When a custom Agent action is not being prioritized correctly by the planner service, the root cause is often missing or improperly defined action dependencies. The planner service determines the execution order of actions based on dependencies defined in the action instructions. To resolve this, the Agentforce Specialist must explicitly specify dependent actions using their API names in the custom action’s configuration. This ensures the planner understands the sequence in which actions must be executed to meet business logic requirements.

Salesforce documentation highlights that dependencies are critical for orchestrating workflows in Einstein Bots and Agentforce. For example, if Action B requires data from Action A, Action A’s API name must be listed as a dependency in Action B’s instructions. The Einstein Bot Developer Guide states that failing to define dependencies can lead to race conditions or incorrect execution order.

In contrast:

Profiles or custom permissions (B) control access to the action but do not influence execution order.

LLM model provider and version (C) determine the AI model used for processing but are unrelated to the planner’s sequencing logic.