Einstein Activity Capture for Sales Cloud Einstein: Let Einstein help keep data between Salesforce and your email and calendar applications up to date. Then you can get back to more important things, like selling.
Activate the MCP Server: Turn on the MCP (Model Context Protocol) Server so Claude can connect to your org.
Add an Action to a Subagent from the Asset Library: Add standard actions to your agent to start handling common business use cases quickly, or create custom actions tailored to your business needs. When you add these actions to an agent, the agent gets its own, independent copy. Changes to template actions within the asset library aren't automatically synced to subagents and actions within your agent. Changes made to one agent's actions don't affect another agent's subagents or actions.
Create a Custom Action: You can create a custom action from the asset library or from within an agent in Agentforce Builder. When you create an action from the asset library, it’s available to add to multiple agents and versions. When you create an action from within an agent in Agentforce Builder, it’s available only to that agent.
Edit the Loading Text for an Agent Action: Loading text tells the user what an agent is doing while an action runs in the background. For editable actions, both standard and custom, you can specify the loading text and tailor it for each action individually.
Explore Standard Agent Actions: Learn all about the agent actions Salesforce provides out of the box in our comprehensive reference library.
Add a Telephony Connection: Add a new Telephony connection to your agent in the Agentforce Builder. The telephony connection optimizes the agent for handling voice calls using your telephony solution.
Choose Your Communication Protocol: Set up Agentforce Voice with PSTN, SIP, Dynamic Routing, or Real-Time Streaming. Setup steps depend on the selected protocol.
Alternative Call Flow for Handling Calls: If your telephony partner creates a third voice call record (VC3) to transfer the call to a human service rep, use this alternative call flow. For example, you might create a third record if you are unable to transfer VC1 to the rep.
Create a Filter to Control Access to Subagents and Actions in the Legacy Builder: Create filters for subagents and actions to make sure that your agent can use them in a conversation only when certain conditions are met. You can limit access based on any variable you specify. For agents on customer channels, we recommend using filters to require authentication for subagents and actions that perform tasks for a user.
Explore the Legacy Agentforce Builder: The Legacy Agentforce Builder is a visual workspace for creating, customizing, and managing agents that support both employee and customer use cases. Use the builder to create agents for helpdesk, deal support, or product assistance and define how they communicate and perform actions. The builder provides a unified workspace for streamlined configuration and testing.
How Agents Work: Follow the path of a user message through the Atlas reasoning engine to learn how Agentforce agents reason, take action, and generate responses.
Access Agentforce Studio and the New Agentforce Builder: The Agentforce Studio app is the home of the new Agentforce Builder and a central hub for building, testing, and monitoring your agents. By default, Agentforce Studio is available in the App Launcher to all users in orgs with access to Agentforce. Each feature within Agentforce Studio requires its own permissions.
Explore Agentforce Builder: Agentforce Builder is where you build, customize, and test predictable, context-aware agents. Learn about the intuitive text editor interface, built-in AI assistance, and more.
Trace Panel in Agentforce Builder: The Trace panel is the reasoning debugger built into the Preview and Live Test experience in Agentforce Builder. It shows how your agent made decisions, including selected subagents, which actions triggered, when topic transitions happened, and how variable values changed.
Variables: To improve security and ensure consistent agent behavior, use variables throughout your agent’s reasoning and interactions, such as in filters, instructions, and logical expressions.
Create Variables in Agentforce Builder: Use variables to store and reuse values in your agent’s reasoning and control how an agent makes decisions. For example, when you assign a variable to an action input, your agent can use only the assigned variable to fill the value of the input.
Configure Call Routing and Call Escalation for the Agent: Configure an Omni-Channel flow to transfer inbound calls to the voice-enabled agent. Create an escalation Omni-Channel flow to disconnect the agent and, if supported by your telephony or CCaaS system, transfer the call to a rep. Add the activated escalation flow to the agent.
Building Agents in Canvas View: Get to know the features of Agentforce Builder’s Canvas view that help you build powerful and reliable agents with Agent Script.
Customize the Persona of Your Agent: Select a persona that you want your agent to use when talking to your customers. Consider the type of persona that is most suitable for your agent’s job or purpose.
Connecting Related Voice Calls: When a customer conversation involves call transfers, a separate voice call record can be created in Salesforce for each call segment. This action results in multiple voice call records for the same conversation. To provide a single, comprehensive view and preserve the full call context, connect these related voice call records. Connecting records enables more informed decisions. For instance, you can escalate the call to the appropriate rep based on all gathered information.
Connect Related Voice Calls When Using PSTN: When an incoming call is transferred to an Agentforce agent using PSTN, a separate voice call record is created in Salesforce for that leg of the call. As a result, you have two voice call records for this conversation with the customer. If you are a Salesforce Voice with Telephony Providers customer, to get a complete view of the entire conversation, connect the Initial and the Agent voice call records.
Connect Related Voice Calls When Using SIP or Dynamic Routing: If you’re a Salesforce Voice with Telephony Providers customer, Salesforce can create multiple voice call records for the same conversation. For example, when an incoming call transfers to an Agentforce agent via SIP or Dynamic Routing, Salesforce creates a separate voice call record for that call leg. If the agent transfers the call to a rep, Salesforce can create a third voice call record. To view the entire conversation with the customer, connect these related voice call records.
Sample Voice Call Connection Configuration in Genesys: The steps for connecting a Rep voice call record (VC3) with an Agent voice call record (VC2) vary for each telephony system. This example illustrates how to perform this configuration in Genesys.
Connect a Service Agent to Partner Telephony: Route voice conversations to your Agentforce Service agents. Learn how to create a telephony connection to your partner telephony system, configure voice mode settings, and route calls to your agent.
Manage Connection Instructions and Response Formats: Guide how your agent reasons and responds on a specific channel with connection instructions and adaptive response formats. For example, tell the agent to ignore disclaimers in emails or to use a specific response format for certain messages.
Set Up Connections in Agentforce Builder: To get started with deploying your agent to channels, add connections to it. Each connection includes behind-the-scenes instructions that help an agent reason and respond for one or more channels. Each connection contains adaptive response formats that help your agent structure responses and deliver multimedia content, such as images, buttons, links, and videos. You can find adaptive response formats and other settings, including the Omni-Channel flows that route conversations to and from an agent, in your connection settings.
Manage Connection Inputs: Connection inputs ensure consistency across agents that use the same connection. Connection inputs can’t be edited during a conversation, ensuring that the values are consistent throughout the agent's interaction. You can customize a specific agent’s connection inputs without impacting other agents.
An Agentforce Guide to Context Engineering: Context engineering is the art and science of giving your AI agent the right information, tools, and instructions to achieve its goals.
Considerations for Conversation Recommendations: Conversation recommendations include two types: welcome recommendations and in-conversation recommendations. Review these considerations before you set up conversation recommendations.
Custom Agent Invocable Action: Example Apex Classes: Call a custom agent invocable action from an Apex class to complete a task automatically with your Agentforce Service agent or default Agentforce agent. The Apex class must return the agent’s response.
Call an Agent from a Flow or Apex Class: Give your users access to your agent’s subagents, actions, and reasoning capabilities to automate repetitive tasks and provide real-time assistance. Use invocable actions to call an Agentforce Service agent, Agentforce Employee agent, or Agentforce (Default) from a flow or Apex class. You can call an agent to complete background or event-driven tasks from anywhere that you can invoke a flow or Apex class.
Example: Improve Your Agent's Memory with Filters and Variables: Help your agent resolve common troubleshooting inquiries effectively with variables and filters. In this example, you create a variable that stores troubleshooting steps related to login issues and filters that control which actions the agent can use. The agent stores all of the troubleshooting steps in the variable. Then, it guides the customer through the troubleshooting process by delivering the most relevant step based on the conversation context.
Agents and Data Usage: As an Einstein customer, it’s important for you to know what data is being used by agents.
Connect an Agent to Slack: Collaborate with Agentforce Employee agents in the same place where your team collaborates. Add the Slack connection to your agent and then finish setting up the agent in Slack.
Connect an Agent to Enhanced Chat v1: Connect an Agentforce Service agent to Enhanced Chat v1 or an Agentforce Employee agent to your Experience Cloud site with Enhanced Web Chat v1.
Connect an Agent to Service Email: Agentforce Service agents can autonomously respond to customer email inquiries. For example, if a customer emails you asking when their package will arrive, your agent can respond to them with the estimated delivery date and tracking number. Email responses are grounded in your Agentforce Data Libraries.
Create an Agent from an Agentforce Employee Agent Template: Agentforce Employee agents help employees find information, complete tasks, and access personalized support across channels. Use Agentforce Employee agent templates to build agents that serve specific departmental needs, support role-based access, and scale securely across the organization. Unlike other agent templates, Employee agents are designed for internal employees, run in the logged-in user context, and you can assign each Employee agent to specific profiles or users.
Connect an Agent to Enhanced Chat: Enhanced Chat is Salesforce’s messaging channel. It allows Agentforce agents to handle customer interactions directly from your website or mobile app. Enhanced Chat currently offers two versions: Enhanced Chat v1 and Enhanced Chat v2.
Use Context Variables in Enhanced Chat Conversations: Personalize agent conversations in Enhanced Chat with context variables. Context variables let your agent use Messaging Session record fields as action inputs.
Personalize Enhanced Chat v2 Interface with Custom Lightning Types: With custom Lightning types, you can enhance the UI of custom agent actions for Agentforce Employee agents in the Agentforce panel in Lightning Experience. This capability applies to agent actions that use Apex as the data type for inputs or outputs.
Example: More Reliable Agent Instructions with Agent Script: Agent Script in the new Agentforce Builder makes it possible to build enterprise-ready agents that get your business processes right every time. This example highlights a few ways Agent Script can help you write instructions that manage instruction overload, reduce latency, and improve overall agent accuracy, instruction adherence, and response quality.
Agent Execution Context and Data Access by Type: Learn how Agentforce Employee, Lead Nurturing (Sales), and Service agents run actions and how data access is controlled for each agent type.
Explore Job-Ready Agents: Get agents working in your business faster with pre-built agents for high-value use cases. These agents are designed to help you get started quickly, with less configuration, a shorter path to production, and faster time to value.
Explore Headless Agentforce Development: Headless Agentforce development is all about building, deploying, and operating Agentforce agents without Agentforce Builder or Setup. Explore resources about how you can author your agents with your AI coding assistant of choice. Learn more about Salesforce skills and other tools to help you through the agent development cycle. Get inspired by end-to-end examples that bring headless tools together to build working agents, no UI required.
Set Up File Uploads for an Agent: Add file upload support to your agent, so your users can share screenshots, receipts, and PDFs. Your agent can interpret individual files or a group of files, answer questions based on one or more files, and add files to a Salesforce record.
Turn Off File Uploads for an Agent: If you no longer want your agent to be able to receive files, turn off file upload capabilities at the agent level.
Instruction Adherence: Instruction adherence is an agent guardrail that evaluates how well your agents interpret subagent instructions when generating responses.
Task Resolution: Task resolution is an agent guardrail. It evaluates how well your agents interpret the user's intent, as expressed in a given conversation, and how well it resolves the task.
Identify a Customer by Phone Number: Give your agent a way to identify a customer by phone number so that voice callers aren't asked to spell out an email address. You build a flow that looks up a contact by phone, wrap it in a custom action, and make that action the agent's only identification path.
Agentforce Language Support: Learn about supported languages, multi-language setup, and agent testing with Agentforce Builder.
Manage Employee Agent Access: Control user access to specific Agentforce Employee agents with permission sets or profiles. In Agentforce Builder, use the Agent Access page. In the legacy Agentforce Builder, use the Agent Access tab.
Manage Conversation Recommendations in the Legacy Builder: You can manage two types of recommendations for an Agentforce Employee agent: welcome recommendations and in-conversation recommendations. Welcome recommendations appear in the Agentforce panel when it opens from most supported record pages, list views, and Lightning app Home pages. In-conversation recommendations appear during an active conversation in response to user messages.
MCP for Agentforce: MCP for Agentforce brings enhanced agent interoperability to Agentforce. Seamlessly register and use third-party tools in Agentforce, so your agents can do more for your customers and employees across systems.
Connect Agentforce Solutions to MCP Servers: The process for connecting Agentforce solutions to an MCP server depends on the source of the server. Connect third-party MCP servers by using the Agentforce Registry. Connect Mulesoft and Salesforce-hosted servers by using API Catalog.
Delete an MCP Server Registration: Delete your server from Agentforce Registry to remove the connection to the server from Salesforce.
Edit MCP Server Details: You can edit a server’s description. If your server uses OAuth 2.0, you can edit the identity provider URL, scope, client ID, and client secret. The server name, server URL, and authentication method aren’t editable.
Manually Apply Policies from an MCP Server Record Page: Apply a policy to your MCP server to control access, set usage limits, and ensure compliance. You can add Agentforce Gateway policies that are available to apply manually.
MCP Risk Score: When you register a third-party Model Context Protocol (MCP) server in Agentforce, Salesforce scans the server's tools and assigns each a risk score. Use the risk score to make informed decisions before allowing the server's tools to access your Salesforce data.
Resolve Out-of-Sync MCP Server Issues: MCP server providers own their servers and can change server and tool settings at any time. Server-side changes make your current registration settings out of sync, which can adversely affect agents that rely on MCP server tools. Agentforce identifies the issue, flags a registered MCP server or tool as Out of Sync, and notifies an admin. Resolve the issue by following the applicable steps.
Add an MCP Tool Action to Your Agent: Add the agent action that references your MCP tool to an agent from the asset library, so your agent can call the tool in conversation.
Delete an MCP Tool Action from the Asset Library: Delete an agent action that references an MCP tool from the asset library so it can’t be added to any agents. When you delete an agent action from the asset library, the action is still available in any subagents and agents you’ve already added it to. We recommend removing the action from any agents before deleting it from the asset library.
MCP Tool Response Schemas and Agent Design: MCP server tools are added to your agent as agent actions that reference the server tool (MCP tool actions) and can be used just like other agent actions. However, not all valid MCP response schemas are fully supported in Agentforce Builder. As you build your agent, plan for differences in variable mapping, action chaining, and input and output rendering. Complex responses may require simpler agent design patterns.
Agent Memory: Agent Memory helps your agents personalize responses by remembering relevant details from your users' conversations. Use it to reduce how often your users repeat themselves and to help agents give more relevant answers over time.
Add a Memory: Add a preference or a detail about your work for an agent to remember, without waiting for the agent to learn it from a conversation.
Delete a Memory: Delete individual memories from your agent's stored memories when they're no longer relevant or accurate.
Edit a Memory: Keep your agent memories accurate by editing stored memory content. Modify a memory to correct inaccuracies, update outdated information, or refine details that agents use in your conversations.
Enable Agent Memory: Enable Agent Memory at the agent level so the agent can recall relevant details from your users' conversations and personalize future responses. Enabling memory applies to all of your agent's users, and for Agentforce Employee agents in Lightning Experience, each user then individually chooses whether to opt in.
Export Memories: Export all stored memories for a specific agent to a markdown file. Use exported files to back up memories or transfer them to another system.
Import Memories: Import memories from a text or markdown file to add memories to a specific agent. Use this to bulk-add memories or restore previously exported memories.
Manage Your Agent's Memories: Manage how your Agentforce agents remember information about you across conversations.
Manage Agent Access to Your Memories: Choose which agents can create memories from your conversations. Turning off access for an agent stops it from creating new memories, without deleting the ones it already created.
Agent Memory Considerations: Review these considerations before you rely on Agent Memory to personalize your agent conversations.
Manage Your Agent's Memories Conversationally: Manage your memory preference and stored memories by talking to your agent, if you're chatting with it outside the Agentforce panel in Lightning Experience.
Manage Your Agent's Memories in Salesforce: Manage your stored memories in the Memory & Preferences panel in Lightning Experience. Choose which agents can create memories, view, search, and filter your memories, edit or delete inaccurate ones, and import or export memories as needed.
Search and Filter Memories: Quickly find specific memories by searching memory content with keywords, sorting by date or agent, and filtering results by agent. Use these options when you have multiple memories stored.
View Stored Memories: Review all memories stored from your agent conversations to understand what information agents use to personalize your interactions. Each memory shows the content, date last updated, and the agent that created it.
Decide Whether to Migrate and When: The question isn’t whether to move to the new Agentforce Builder. It’s when to get started and what to bring with you.
Implementing Your Migration: With your tools selected and your Agent Script improvement plan in place, it’s time to migrate. Take a staged and iterative approach to make your migration easier to manage, test, and troubleshoot. Get tips for working effectively with testing tools, third-party AI tools, and your team.
Agentforce Agent Migration Guide: If you’re an Agentforce customer using agents you built in our legacy builder, this guide is for you. It’s a centralized resource for migrating your existing agents from the legacy Agentforce Builder in Salesforce Setup to the new Agentforce Builder in the Agentforce Studio app. It helps you decide if, why, and how to migrate, and it covers the tools and guidance you need to make your migration cost- and time-effective. Use it to go from a prompt-based, sometimes unpredictable agent to a precise, reliable agent that reasons when it has to and just gets things done when it doesn’t.
Planning and Designing for Agent Script: Before migrating, spend time designing your subagent instructions and conversational flow. The level of time and effort varies depending on what tools you plan to use to migrate, the complexity of your agent, and how well your agent performs today. This prep helps you determine what parts of your agent can be migrated directly with no changes and what parts are candidates for redesign. It also prepares you to make deliberate decisions about where to introduce determinism in your new agent, so you can improve your agent strategically instead of reactively.
Preparing to Migrate: Before you begin, familiarize yourself with the new Agentforce Builder and Agent Script, learn how the upgraded reasoning engine works, and explore migration tools to choose the approach that works best for you.
Monitor Agent Health: Agent Health Monitoring detects and notifies you of silent agent failures—like spiking error rates or high latency—in near-real time across your AI agents. Rapidly respond to agent performance issues to mitigate customer impact.
Manage Alerts for Agent Health Monitoring: Set up an alert to monitor a metric for a particular agent. You receive an email notification when a performance data breaches an alert threshold.
Investigate Notifications: Manage the notifications that you received from monitoring alerts that you created. The system sends a notification to your email inbox whenever agent data for a monitored metric breaches an alert threshold.
Set Up Agent Health Monitoring: Set up Agent Health Monitoring so you can create alerts and get notified when agent performance breaches alert thresholds. Observe agent health status and trends. Rapidly investigate and resolve agent performance issues.
Multi-Agent Orchestration: Extend your agent’s capabilities with Multi-Agent Orchestration for Agentforce. Connect your Agentforce agents with other specialized Agentforce agents in your Salesforce org so that they can collaborate securely on complex tasks. With multi-agent orchestration, businesses can deploy a single agent across their channels with shared context and without customer friction.
User-Level Authentication in Multi-Agent Solutions: Agentforce multi-agent solutions support user-level authentication. Authentication varies by agent type and by the combination of orchestrator agent and connected subagent.
Connect an Agent as a Subagent: Add an agent as a connected subagent to your orchestrator agent. Your orchestrator agent can coordinate efforts among connected subagents, such as delegating tasks and synthesizing results.
Remove a Connected Subagent from an Orchestrator Agent: Remove a connected subagent you no longer need from your orchestrator agent. Removing the connected subagent doesn’t affect the reference agent on which this connected agent is based.
Build, Extend, and Troubleshoot with Agent Optimizer in Agentforce Builder (Beta): Focus on what you want your agent to do, not how to configure or fix it. Agent Optimizer is a built-in coding agent that extends, refines, and troubleshoots your agents directly in Agentforce Builder, at no additional cost. Explore use cases and considerations for working with Optimizer.
Configure Your Agent: Customize your agent so that it fits your company’s use cases. Learn how to use Agentforce Builder and agent assets, including subagents, actions, filters, and variables, so that you can customize out-of-the-box solutions or create your own.
Ground Agentforce in Data: A great agent uses your data to provide high-quality responses. Learn about methods for connecting agents to knowledge sources, including internal data, unstructured documents, and web search.
Deploy Your Agent to Channels: Connect your agent to multiple channels. Meet your customers and employees where they spend the most time.
Connect an Agent to Messaging: Route conversations to your Agentforce Service agents and Agentforce Employee agents with messaging channels, including Enhanced Chat and other enhanced messaging channels. Learn how to use context variables, multiple languages, and progess indicators in messaging channels.
Extend Your Agentforce Implementation: Explore more advanced use cases with the help of our API and Developer guides. You can also call an agent from a Flow or Apex class.
Ideate Your Agentforce Solution: Build your Agentforce solution from the ground up by understanding the basics of a successful AI project. Then, learn about the basics of agents so that you can get started quickly.
Monitor Your Agent: When you review and update your agentic solution, it has a higher chance of success. Review Agentforce analytics to report on key indicators such as performance, usage, quality, trust, cost, and ROI.
Set Up Your Agent: Before you launch agents, it’s important to understand what needs to happen first. Understand key prerequisites, turn on Einstein and Agentforce, and prepare your data.
Test Your Agent: Testing begins with single-session testing in Agentforce Builder, where you can manually validate agent behavior and troubleshoot issues in detail. When the agent performs reliably in individual scenarios, testing can scale by using batch tests in Agentforce Testing Center. Batch testing provides a broader view of performance across diverse scenarios, helping identify patterns, edge cases, and gaps that didn’t surface during manual testing. Together, these approaches make sure that the agent is well-tested, consistent, and ready for real-world use.
Variables: With variables, your agent can store and reuse values that help it with reasoning and interactions. Variables work differently in Agentforce Builder and the legacy builder.
Plan Your AI Agent: Planning an AI agent is crucial for several reasons. It ensures that your AI project aligns with your company's overall AI strategy and business goals. It also helps you define the agent's work and assess its potential impact on the organization.
Design the Subagents and Actions: After identifying the right subagents and actions for your AI agent, start planning how to build them. Following best practices when designing the subagents and actions can help your agent perform reliably and effectively.
Identify the Subagents and Actions: After determining your agent’s use case and the work the agent will do, you can identify the right subagents and actions for the agent. Agents are made up of subagents, which define the different jobs an agent can do. Subagents also contain a set of actions, which are the tools the agent can use to do its job.
Salesforce Considerations for Agentforce Projects: Your AI agent must coexist with your current Salesforce configuration and architecture. Consider Salesforce-specific factors that can influence your Agentforce implementation.
Map Business Processes for an Agentforce Project: When planning an AI agent, it’s important to define the work the agent will do. One way to define the work is to map the business processes related to your Agentforce use case.
Define the Technical Requirements: When planning an AI agent, gather technical requirements that will help you build the right solution. Key factors to explore include: data, channels, routing and escalation, and security controls.
Identify Risks and Guardrails for an Agentforce Project: Autonomous AI agents come with risks, such as security threats, data breaches, reputational harm, financial loss, bias, hallucinations, and issues with transparency and accountability. When planning an AI agent, identify risks for your Agentforce use case and plan risk mitigation strategies.
What is an Agentforce Use Case?: An Agentforce use case is an application of AI technology where an AI agent takes an action or series of actions that accomplish a goal or job-to-be-done on behalf of your employees, customers, or organization.
Define an Agentforce Use Case: After identifying your use case ideas for Agentforce, flesh out each of the use cases so your organization can assess and prioritize them. Remember, this stage of the planning process is about the project goals, not the technical solution.
Agentforce Builder Testing: Manual testing, like you do in Agentforce Builder, is the starting point for validating your agent’s performance. It offers a detailed, step-by-step evaluation of your agent's interactions and utterances and is especially effective for confirming configurations and exploring how the agent responds to different inputs. Throughout the agent lifecycle, plan to revisit manual testing in Agentforce Builder, especially after making changes to your agent.
How Agents Created in the Legacy Builder Work: Follow the path of a user message through the Atlas reasoning engine to learn how an agent built in the legacy Agentforce Builder reasons, takes action, and generates responses.
Recommended Call Flow for Handling Calls: The recommended call flow minimizes the number of voice call records and connects related records. When possible, use this flow for handling the different legs of the call.
Turn on Voice Call Recording and Transcription When Using SIP or Dynamic Routing: Turn on voice call recording to record calls between customers and agents. Recordings use stereo separation for easy speaker identification, placing the customer on one channel and the agent on the other. Salesforce stores each recording as an MP3 file and links it directly to its voice call record.
Turn on Voice Call Recording When Using PSTN: To record the conversations between customers and Agentforce agents, turn on the recording. By default, the recording is turned off.
Turn on Voice Call Recording and Transcription When Using Real-Time Streaming: To have Salesforce record the voice calls between customers and the Agentforce agent, turn on the voice call recording. To easily distinguish speakers during playback, recordings are stereo-separated, with the customer on one channel and the agent on the other. Each recording is stored as an MP3 file and automatically linked to its voice call record for easy access.
Security | Anomaly Details: Provides detailed analysis of security anomalies, including user activity patterns and security implications.
Automotive | Check Asset For Warranty: Gets the information for a valid asset warranty for a specific part by using its serial number, or through its root asset, such as a vehicle or a battery, if available.
Automotive | Check Coverage for Parts: Validates and retrieves serializable parts within an asset's hierarchy. If none are found, retrieves the non-serializable parts with matching names covered by the main or the root asset's warranty terms.
Automotive | Create Asset Appraisal: Creates an appraisal record for an asset and an appraisal item associated with it based on an input that defines the trade-in and opportunity context for the flow, capturing the opportunity ID, the specific record ID of the customer's trade-in asset, and the condition status of that asset.
Automotive | Create Opportunity with Products and Preferred Seller: Creates an opportunity, associated opportunity products, and opportunity preferred seller records for a customer based on inputs that define the core record identifiers and data collections for the flow, including the account, product, and preferred dealer IDs, alongside an Apex collection containing the pricebook mapping details required for opportunity creation. Captures the details of the newly generated records, including the opportunity's core attributes, the associated preferred seller record, and specific line-item data such as the opportunity ID, price book entry, product ID, quantity, and unit price.
Automotive | Create Quote from Opportunity: Creates quotes and quote line items for an opportunity using its ID as an input and its associated opportunity products with a manual discount percentage.
Automotive | Get Pricebooks From WorkType: Gets the product IDs associated with a worktype and subsequently gets the pricebook records containing all the retrieved product records.
Automotive | Get Warranty Claims for Account: Gets a list of warranty claims associated with a specific account, using optional parameters such as an asset ID, an asset serial number, or a claim number.
Automotive | Search Customer Accounts: Validates the agent session and searches for customer accounts by using a provided email address, phone number, or account name.
Automotive | Search Dealers for Products: Gets the dealer product searchable field records based on location and user specified search criteria and filters the records to find dealers that offer user requested products in a service territory.
Automotive | Search Product And Parts: Gets products and associated parts or accessories based on inputs that define the search parameters, including the text-based query string or the base product ID, while using boolean flags to distinguish between text searches and product-related component searches. Provides a list of matching products with their respective pricing, or prompts users for more specific search queries.
B2C Commerce | Answer Product FAQs: Answers frequently-asked questions about B2C Commerce store products to help customers make informed buying decisions.
B2C Commerce | Answer Site FAQs: Answers frequently-asked questions about you B2C Commerce storefront. This action generates answers from files about your site, stored in the in the Agentforce Data Library.
B2C Commerce | B2C Commerce Product Search: Provides category recommendations, product recommendations, or query reformulations if the customer's request shows an intent to buy, explore, or learn about products that are sold in the store.
B2C Commerce | Get B2C Order Confirmation Details: Retrieves order confirmation information based on the order ID. Use this action to view order confirmations after a successful payment for a new order. It's invoked automatically. Don't use it to search for a shopper's order.
B2C Commerce | Get B2C Product By ID: Retrieves product information based on a Product ID. Information retrieved includes basic product details, along with related information such as product variations, and product bundles.
Unified Catalog | Check Product Eligibility: Checks whether a product is eligible for a contact, an account, or a user based on predefined criteria. The action can verify product eligibility when accounts, contacts, or users self-serve, or when service reps or request fulfillers assist requestors.
Collections and Recovery | Compile Script for Next Call Based on Previous Collections Interactions: Gets recent voice call records and their transcripts, truncates the combined transcript text to a predefined character length, and provides the summary to the user. Also, gathers notes associated with the collection plan, combines note titles and text previews, truncates this combined text to a predefined length, and provides the summary to the user. Compiles a call script for collections specialists for their next customer conversation.
Collections and Recovery | Create Case and Related Case Proceeding Records: Creates a case for the collection plan by using the details provided by the user. If the case type is Legal, the flow verifies if the case is eligible for legal proceedings. If it does, then it creates the related case proceeding and case proceeding participant records.
Collections and Recovery | Create Promise to Pay: This action lets collections specialists create promise-to-pay agreements for borrowers. The action gets relevant collection plan record details for a collection plan ID. Gathers the payment commitment details from the user and creates a payment schedule and payment schedule item records, and sends an email to the user with the promised payment details.
Collections and Recovery | Get Collection Plan Details by Name: Gets a collection plan record details by using the collection plan name, and verifies if it's related to the querying customer. If related, returns the collection plan details in a formatted string format.
Collections and Recovery | Get Collection Plans for Related Contact and Parties: Gets a specified number of collection plans associated with a customer's contact. If required, it checks linked financial accounts for additional collection plans until the limit is reached, compiles the details, and returns them to the customer.
Collections and Recovery | Generate and Send Payment Link: This action creates a payment link by using the payment details provided by the user. If an email ID is available, sends an email to the borrower with the payment link details. Otherwise, shows the payment link details to the user who triggered the action.
Collections and Recovery | Request Payment Extension: This action gets relevant collection plan record details for a collection plan name. Gathers the payment commitment details from the customer and creates a payment schedule and payment schedule item records, and sends an email to the customer with the agreed payment details.
Collections and Recovery | Update Representative Contact for Collections: This action gets relevant collection plan record details for a collection plan ID. Checks if the representative contact is already available. If it's available, notifies the customer that the new representative's details can't be updated. If representative contact isn't available, gathers the details from the customer, creates a new contact record using these details, and updates the collection plan record accordingly.
Communications | Get Add-On Products: Gets a list of compatible add-on products for a customer’s existing asset, including associated pricing data. This action requires the asset ID and pricebook ID for successful completion.
Communications | Get Cross-Sell Products: Gets a list of active cross-sell product suggestions, including pricing, for a specified account, catalog, and category.
Communications | Create Billing Dispute Case: Creates a case to raise a billing dispute after running the Get Account Record, Get All Invoices, Get Invoice Details, and Get Invoice Charges actions. Check the invoice number and charge for the line item in the invoice that requires a case. After the case is created, the Case Record URL is shown in the conversation.
Inventory Management | Create Product Request: Create product requests for specified products and quantities from the given location to address shortage scenarios. This action supports the creation of product requests to streamline inventory management.
Service | Create a Sales Contact Record: Searches for a contact record that matches the information provided by a prospect during a web inquiry. If one doesn’t exist, a new contact record is created using the information provided.
Service | Create a Sales Lead Record: Searches for a lead record that matches the information provided by a prospect during a web inquiry. In one doesn’t exist, a new lead record is created using the provided information.
Service | Create a Sales Person Account Record: Searches for a person account record that matches the information provided by a prospect during a web inquiry. If one doesn’t exist, a new person account record is created using the information provided.
Sales | Create a Sales Record: Compares the customer’s entered code with the generated verification code from the Send Verification Code agent action to determine if the customer is verified.
Sales Agreement | Create Visit: Create a visit based on user specified instructions and dates for a specific context, such as a sales agreement.
Digital Lending | Analyze Estimated Loan Amount: Determines the estimated loan amount based on the monthly payment and loan tenure you provide. The action uses the product list rate of the specified product or a nominal rate to calculate the annual interest rate.
Education Cloud | Create Inquiry: Creates a case, an educational info request, and an academic interest raised by prospective students, parents, or counselors. Call this action when there isn’t an answer to the user’s question.
Education Cloud | Get Account Balance: Retrieves information about the total fee balance for a student using the date and optional fee type provided as inputs. The action uses the date to determine which academic term to retrieve fees for.
Education Cloud | Get Attendance Info: Gets attendance information from an incoming voice call, including the attendance recordings, contact IDs, source types, source data timestamps, and formatted entry details for absent students.
Education Cloud | Get Fee Breakdown: Retrieves the details of how a fee was calculated for a specified order details ID and order item ID. The action explains what the fee covers and any policies relevant to the calculation.
Education Cloud | Get Fee Details: Retrieves the details of how a fee was calculated for a specified order details ID and order item ID. The action explains what the fee covers and any policies relevant to the calculation.
Education Cloud | Suggest Student Goals (Beta): Generates personalized, actionable goals for a student based on their profile, preferences, and assessments, supporting their personal growth, skill development, and career readiness.
Energy & Utilities | Generate Tariff Pricing Request: Generates the pricing request for the Get Fully Priced Products action from the selected fuel category, consumption context, chosen service points, and price-impacting tariff attributes. Emits one service point entry per required fuel (Electricity, Gas, or both for Dual Fuel), each with its consumption details and resolved tariff attributes.
Energy & Utilities | Get Available Service Points: Gets a collection of available service point records at the selected premise, including market identifiers and distribution regions, to drive category selection, pricing, and enrollment.
Energy & Utilities | Get Customer Service Addresses: Gets a collection of available service point records at the selected premise, including market identifiers and distribution regions, to drive category selection, pricing, and enrollment.
Energy & Utilities | Get Energy Products: Gets a list of products and their list prices from a price book in the Energy & Utilities product catalog that match the specified filter criteria.
Energy & Utilities | Get Tariff Configurations: Gets the tariff configuration for a specified catalog and category to return the catalogs, categories, consumption parameters, and tariff attribute schema.
Energy & Utilities | Submit Customer Enrollment: Submits a tariff enrollment order for a known Energy & Utilities customer using the provided customer details, pricing details, and service points.
Field Service | Get Current Time for Field Service: Gets the current time according to the time zone. The time zone is taken from the operating hours that are associated with the service appointment’s service territory.
Field Service | Get Engagement Topics for Field Service: Retrieves an engagement topic’s ID and object name for the current messaging session. The engagement topic is used to connect between a messaging session and a service appointment.
Field Service | Rebook and Notify for Field Service: Notifies the customer that their service appointment needs to be rescheduled. This action is part of Customer Outreach, which is an Agentforce Scheduling feature.
Sales | Find Potential Prospects: Finds potential prospects based on job titles and account context. Searches Salesforce records and other sources for job titles at different seniority levels and returns a list of potential prospect information.
Financial Services | Generate Resolution Note: Generates a concise, factual, and audit-ready resolution summary for internal complaint records by using the provided structured data.
Financial Services | Create Case for Transfer Funds: Creates a case to transfer funds between the customer’s financial accounts. Supports transfers between internal-to-internal accounts, external-to-internal accounts, and internal-to-external accounts, including Own Account Transfers, Loan Payments, and Credit Card Payments.
Financial Services | Create Financial Goal: Collect the account ID, start date, status, type, target amount, priority, target date and name. Create a financial goal for the account.
Financial Services | Create Interest Tag: Extract interest tags from the meeting notes. Show the extracted interest tags to wealth management advisor, who can then select the interest tags to create.
Financial Services | Extract Action Plans: Extract action plans from meeting notes. Identify one or more action plans, statuses, and start dates for wealth management advisors.
Financial Services | Extract Financial Goals: Extract financial goals from meeting notes. Identify financial goals, types, statuses, priorities, start dates, target amounts, and target dates for wealth management advisors.
Financial Services | Extract Financial Plans: Extract financial plans from meeting notes. Identify financial plans, types, statuses, priorities, start dates, target amounts, and target dates for wealth management advisors.
Financial Services | Fulfill Card PIN Reset: Submits the card details, reset reason, and selected method to the network for verification. If successful, creates a closed case and resolves the request, else returns an error message.
Financial Services | Get Card Settings: Gets the details of the preconfigured settings associated with the card selected by the customer to raise a request.
Financial Services | Get Check Details: Gets the status and details of an issued check associated with the financial account and check number provided by the customer.
Financial Services | Get Complaint Subtypes: Retrieves valid complaint subtype values for the selected complaint type. Returns a sorted list of applicable complaint subtypes based on the selected complaint type.
Financial Services | Get Complaint Types: Retrieves valid complaint type values for the selected product type. Returns a sorted list of applicable complaint types based on the selected product type.
Financial Services | Get Financial Accounts for Complaint: Retrieves financial accounts associated with the customer, filtered by the selected product type. Returns account details for the customer to select the account related to their complaint.
Financial Services | Get Transaction Dispute Questionnaire: Retrieves the assessment questionnaire based on reason code, reason subcode, and payment network for disputed transactions. Generates an assessment ID after the questionnaire is completed and updates the assessment record with the case details.
Fundraising | Close Gift Commitment: Closes a gift commitment by updating the status of all unpaid and failed transactions to Closed. Returns a confirmation that the gift commitment will be closed on the specified date.
Fundraising | Manage Recurring Gift Commitment: Pauses, resumes, or updates the amount on a recurring gift commitment. Returns a confirmation that the requested action was successful or encountered an error.
Fundraising | Prepare Donor Meeting Notes: Transform raw donor meeting notes into a structured summary with key discussion points and next steps. Use when a major giving officer provides notes or a recap from a donor meeting.
Marketing | Generate Campaign Insights: Retrieves engagement metrics, such as open rates, clicks, or unsubscribes for a specific campaign. It then generates key content insights about the message activities from the campaign's first flow. For example, identify which channel had better engagement or lower engagement over time.
Revenue | Generate Product Description: Generates product descriptions within Product Catalog Management in Revenue Cloud. This action uses product record data to create clear and engaging descriptions, enhancing product discoverability and expediting catalog creation.
Communications | Get Account Balance: Fetches and shows the balance and service details for a given account number whenever the user requests their account balance.
Communications | Get Account Record: Fetches an account record using the account number and contact email address that’s associated with the account.
Sales | Get Fields to Create Sales Record: Collect the fields that need user-inputted values in order to create a sales record. Also outputs if the agent should create a lead, contact, or person account record.
Communications | Get Invoice Details: Fetches and shows key invoice details, including invoice amount, billing date, due date, and payment date, for a given invoice number provided as input.
Sales | Get Record Research: Gathers research records from the last 30 days that are associated with a specific record ID. Then, the action provides a summary of the retrieved insights and research data for that record. This action is used within Agentforce Sales Management.
Sales | Get Related Emails: Gets a list of the latest related email records within a date range, based on a user’s query. This action is used within Agentforce Sales Management.
Sales | Get Related Notes: Gets a list of the latest related note records within a date range, based on a user’s query. This action is used within Agentforce Sales Management.
Retail Execution Visit Assistant | Get Retail Execution Account Performance Data: Calculates total revenue, minimum and maximum order values, and identifies top N% and bottom N% selling products based on order quantities. The action calculates product metrics data from preorders, deliveries, and invoices that are ready or released, optionally within a date criteria.
Retail Execution Visit Assistant | Get Retail Execution Orders By Account: Gets order information, including order header details and corresponding detailed order item data for each order. The action gets ready or released preorders, deliveries, and invoices data for the account, optionally within a date range.
Retail Execution Visit Assistant | Get Retail Execution Orders By Visit: Gets order information, including order header details and corresponding detailed line-item (order item) data for each order. The action gets ready or released preorders, deliveries, and invoices data for the visit, optionally within a date range.
Retail Execution Visit Assistant | Get Retail Execution Visits By Account: Gets completed visits for the account based on the preconfigured visit types in Retail Execution. The action supports two scope modes: returning only the most recent completed visit or returning all completed visits within a specified date range. Each retrieved visit includes associated details of store audits, product surveys, and asset audits.
Security | Get Security Metric Types: Retrieves all the Security Center metric type enums and their descriptions for mapping metric types to readable names.
Automotive | Get Updated Terms for Payment Deferral: Sends a request to check the eligibility of a payment deferral request for a loan, a lease, or any other financial product to an external service, and returns the updated payment terms.
Automotive | Get Updated Terms for Payment Due Date Modification: Sends the request to check the eligibility of a payment due date modification request for a loan, a lease, or any other financial product to an external service, and returns the updated payment terms.
Health | Add Patient to Waitlist: Adds a patient to a waitlist for a specific work type and service territory, with an optional preferred provider. Returns whether the patient was added, an error message if the action fails, and the new waitlist participant ID.
Health | Book Appointment: Books an appointment for a patient in a selected slot with a provider or asset, for the given work type and time. Returns the created appointment ID and status.
Health | Summarize Disease Surveillance Case: Summarizes a patient’s case, including investigations, lab results, risk assessments, symptom onset, hospitalization status, and any interventions given.
Health | Classify Disease Surveillance Cases: Suggests a classification for a disease investigation case as Confirmed, Suspected, or Probable based on the lab findings and the criteria outlined in the disease definition.
Health | Create New Patient: Creates a PersonAccount record in Agentforce Health for a patient who isn't yet in the system. Collects first name, last name, date of birth, and phone number. Doesn't collect email or address by voice.
Health | Summarize Disease Investigation Details: Summarizes a disease investigation, including reported symptoms and onset date, possible exposure sources such as travel history and recent contacts, and any relevant inspection findings or other investigative details.
Health | Extract Patient Home Visit Requirements: Gets the transcript of the patient's voice call with the home health agency and extracts details about home visit requirements, benefits, and quotes. Drafts a rich-text summary of the patient's home healthcare requirements and creates a JSON response with the patient’s quote and benefit information.
Health | Fetch Individual Record ID: Retrieves the Individual ID from Data 360 by using member information. Accepts either a member ID or the member’s first name, last name, and date of birth.
Health | Fetch Prior Authorization: Retrieves prior authorization (care request) records for a member based on prior authorization ID, claim ID, or member and service details.
Health | Find Claims: Retrieves claim records for a member from an external zero copy system based on the member number, service start date, and service end date. Optionally, filters by claim status and provider name. Returns up to the 5 most recent claims, sorted by claim creation date.
Health | Find Claim Clinical: Retrieves clinical information for a specific claim from an external zero copy system based on the claim ID. Returns diagnosis records, supporting information, patient health conditions, and medical procedures.
Health | Find Claim Errors And Cases: Retrieves error and case information for a specific claim from an external zero copy system based on the claim ID. Returns care processing errors and claim case records.
Health | Find Claim Payment: Retrieves payment information for a specific claim from an external zero copy system based on the claim ID. Returns claim payments, coverage records, and payment detail breakdowns.
Health | Find Claim Provider: Retrieves provider information for a specific claim from an external zero copy system based on the claim ID. Returns healthcare provider details, NPI records, and facility information.
Health | Find Member: Retrieves the member number from Data 360 by using a combination of first name, last name, and date of birth.
Health | Find Provider Claims: Retrieves claim records for a healthcare provider from an external zero copy system based on the provider NPI, service start date, and service end date. Optionally, filters by member number, claim status, preauthorization ID, or claim ID. Returns up to the 20 most recent claims.
Health | Find Specific Member Coverage: Extracts specific member coverage details based on user context and requests. Queries member coverage data and returns only the specific information requested by the user.
Health | Generate Member Search Payload: Generates the payload for searching the member details based on the user's utterance. Summarizes the JSON generated by the Generate Member Search Payload prompt template.
Health | Generate Provider Directory Search Payload: Generates the payload for searching the provider directory based on the user's utterance. Summarizes the JSON generated by the Generate Provider Directory Search Payload prompt template.
Health | Generate FHIR Payload: Generates the FHIR input payload required to register a patient in the Electronic Health Record (EHR) system.
Health | Generate Home Healthcare Post-Visit Summary: Creates a post-visit summary of a patient's home healthcare visit based on the associated work order. The summary includes post-visit details, previous visit conclusions, and comprehensive 360° patient data, such as care activities, observations, vitals, medications, and conditions. The summary also covers follow-up actions captured as work steps and tasks in the work order and the patient's clinical history.
Health | Get Active Appointment Reasons: Returns the list of active appointment reasons available for patient self-scheduling. Provides readable text that the voice agent can read to patients when they ask which reasons are available, when they want to repeat them, or when a direct work type match fails.
Health | Get Appointment Slots: Retrieves available appointment slots for a specific provider or asset based on the work type and requested date range.
Health | Get Channels For Work Type Id: Retrieves the active Salesforce Scheduler engagement channel types configured for a given WorkType through the EngagementChannelWorkType junction. If any junction row for the work type supports all channels, returns every active Salesforce Scheduler channel. Otherwise, returns only the linked active Salesforce Scheduler channels. Use this action during appointment booking to find available channels for selection.
Health | Generate Claim Summary: Provides a summary of a claim. For bulk claim queries, shows a snapshot of up to the 5 most recent claims.
Health | Get Input for Eligibility and Benefit Verification: Gets the eligibility and benefit verification request input ready, including payerIdentifier, providerIdentifier, and facilityIdentifier to retrieve the member plan details based on the provided payer and service information.
Health | Get Member Coverage Summary: Retrieves and summarizes member coverage information. Queries member coverage and provides a comprehensive summary of the member's coverage.
Health | Get Patient Data from EMR: Gets the clinical, demographic, and administrative patient data from the Electronic Medical Record (EMR) system.
Health | Get Patient Member Plans: Gets the MemberPlans of the patient from Salesforce. If no results are returned from this action, then invoke the GetPatientDataFrEmr action with the patient's details.
Health | Get Patient Referrals: Retrieves active accepted referrals for a given patient. Returns all matching referrals and WorkType details.
Health | Get Provider NPI: Takes a provider ID as input and returns the National Provider Identifier (NPI) for that provider.
Health | Get Provider NPI By Account: Retrieves the National Provider Identifier (NPI) from CRM objects by using a provider Account ID to resolve the associated Healthcare Provider record.
Health | Get Specialties For Work Type: Retrieves the list of care specialties configured for a given WorkType through the WorkTypeCareSpecialty junction. Returns only active CareSpecialty records. Use this action during appointment booking when a patient has no assigned provider to get required specialty criteria for a provider search.
Health | Link Verified Account: Links the verified patient account to the active voice call or messaging session. This background step runs automatically after verification succeeds.
Health | Summarize Patient Healthcare Details: Creates a rich-text summary of a patient’s healthcare details based on the patient name. The summary includes patient information such as opportunities, clinical service requests, and member plans.
Health | Summarize Patient Home Healthcare Details: Creates a rich-text summary of a patient’s home healthcare details based on the patient name. The summary includes patient information such as opportunities, clinical service requests, and member plans.
Health | Verify Patient Scheduling Eligibility: Verifies whether a patient appointment scheduling request can be handled by the AI agent, or must be routed to a human representative based on EHR system integration and scheduling policies. This action evaluates patient eligibility for automated scheduling in the context of external EHR system workflows.
Security | Identify Anomalies: Analyzes security alerts and metrics to identify anomalies and potential threats for each tenant and metric type.
Marketing | Identify Business Unit: Retrieves engagement metrics, such as open rates, clicks or unsubscribes, for a specific campaign and generates key insights about the campaign’s performance. For example, recommend different subject lines or add more time between email sends to prevent subscriber fatigue.
Security | Investigate User: Investigates a specific user's activity, reports, object access, and permission path to help analyze anomalies and answer investigative questions.
Agentforce IT Service | Add Comment to Record: Adds comments or supplementary information to existing IT tickets through a conversational interface, streamlining the update process by allowing users to identify tickets by number or description without manual navigation.
Agentforce IT Service | Answer Policy Question: Provides employees with personalized, context-aware answers to questions about company policies, benefits, and entitlements through a conversational interface. This action uses profile data, such as role, location, and tenure, to summarize direct answers from policy documents and provide concise responses.
Agentforce IT Service | Answer Questions with Knowledge: Resolves user inquiries by referencing a knowledge base of company policies, troubleshooting guides, and product documentation. Use when a user asks about return or refund policies, how to diagnose or fix an issue, or what capabilities a product offers.
Agentforce IT Service | Associate Related Records For Incident: Finds similar or related records for an incident and creates associations, such as problems or change requests, to an incident. The action finds other IT records that are either similar or caused by the source incident and creates associations automatically.
Agentforce IT Service | Associate Related Records for Problem: Finds similar or related records for a problem and creates associations, such as incidents or change requests, to a problem record. The action finds other IT records that are either similar or caused by the source problem and creates associations automatically.
Agentforce IT Service | Check Incident Attributes: The Check Incident Attributes agent action checks and gets important information about an incident, like its status, priority, and assignment. This information is important for managing and solving the incident.
Agentforce IT Service | Check Problem Attributes: Automatically validates attributes such as impact, urgency, category, subcategory for a problem to check whether they’re correctly assigned. This action uses historical data and record context to suggest corrections, reducing manual errors and resolution delays.
Agentforce IT Service | Create Incident Resolution Summary: Creates a resolution summary for an incident based on related and similar records, feed and swarm data, knowledge articles, and activities. The action reduces manual documentation effort, ensures consistency, and creates valuable grounding data for use in future knowledge articles or other agent actions.
Agentforce IT Service | Create Incident For Employee: Creates an incident record for an employee issue when the problem can't be resolved through existing knowledge articles or no relevant articles are available. The action reduces friction in the escalation process and ensures unresolved issues are formally tracked.
Agentforce IT Service | Create Incident Root Cause Summary: Creates a root cause summary for an incident based on related and similar records, past incidents, configuration items, knowledge articles, and activities. This helps incident managers understand and record the reasons for the issue, serves as grounding data for other agent actions, and helps the team prevent similar issues in the future.
Agentforce IT Service | Create Problem Resolution Summary: Creates a resolution summary for a problem based on related and similar records, feed and swarm data, knowledge articles, and activities. The action reduces manual documentation effort, ensures consistency, and creates valuable grounding data for use in future knowledge articles or other agent actions.
Agentforce IT Service | Create Service Catalog Request: Initiates and submits hardware service requests through a conversational interface. This action validates required inputs, prompts for missing information, and confirms request details to streamline the procurement process.
Agentforce IT Service | Create Problem Root Cause Summary: Creates a root cause summary for a problem based on related and similar records, past problems, configuration items, knowledge articles, and activities. This helps problem managers understand and record the reasons for the issue, serves as grounding data for other agent actions, and helps the team prevent similar issues in the future.
Agentforce IT Service | Delete Configuration Item: Performs a soft delete of a configuration item (CI) in the CMDB based on a CI ID. The action marks the CI as deleted without permanently removing it from the database. Before calling this action, call Get Configuration Item Details to display the full CI details to the user and capture explicit confirmation. Use the CI ID returned by Get Configuration Item Details as the input to this action.
Agentforce IT Service | Employee Incident Creation: Creates an incident for employee issues from Agentforce conversations, if they can't resolve the issue with knowledge articles or if no articles exist. This action automatically creates incidents for employee issues and saves their time and effort.
Agentforce IT Service | Evaluate Problem Creation Requirement: Evaluates whether a problem must be created or not for an incident based on existing associated problems, similar historical issues, and the incident's severity. This helps to further investigate the issues and reduces manual decision-making.
Agentforce IT Service | Execute Service Catalog Item Flow: Runs a service catalog item flow based on the user's input. The action initializes the flow using a product ID, then processes subsequent interactions using the interview ID and user input until the flow reaches a finished or fault status.
Agentforce IT Service | Generate Post-incident Review: Generates a post-resolution summary for an Incident record using details such as the category, subcategory, description, related records feed, knowledge, and swarm data into the summary. This action is triggered when an incident fulfiller requests a summary of an Incident in a closed or resolved state.
Agentforce IT Service | Get Assigned Software For Employee: Retrieves the list of software applications currently assigned to the logged-in user from Okta. The action uses the current user's email to look up their Okta profile and get the assigned app links.
Agentforce IT Service | Get Available Software For Employee: Searches and retrieves a list of available software applications based on the user's search query. The action understands the intent of the user's input to return the most relevant software options and generates a summary of the matching results.
Agentforce IT Service | Get Configuration Item Details: Retrieves details for a specific configuration item (CI) based on the user's input, such as a CI ID or CI name. The action identifies the requested CI and returns key attributes including name, class, status, and owner. Use this action when users request CI details or need to perform update or delete operations on a specific CI.
Agentforce IT Service | Get Current Assigned Asset Details: Retrieves detailed information about assets currently assigned to the authenticated user, enabling employees to quickly view their allocated IT assets without navigating through multiple systems or submitting manual lookup requests.
Agentforce IT Service | Get Eligible Service Catalog Items: Retrieves service catalog items that match a user's request by analyzing their description and returning only the options they are eligible to access. The action reduces time spent manually browsing the catalog, and surfaces the most relevant service catalog items quickly.
Agentforce IT Service | Get Incidents Based on Description: Retrieves a list of records based on the user's description for specific types of tickets such as cases, incidents, or service requests. The action uses AI-powered filtering to get the most relevant list of records by understanding the intent of the user input and generates a summary of the relevant records.
Agentforce IT Service | Get Product Launch Card: Retrieves the launch details for a product, including a cover image and a direct link to the relevant intake form. The action reduces time spent searching for the correct entry point, ensures users are directed to the right form quickly, and provides a consistent starting point for use in intake workflows.
Agentforce IT Service | Get Service Catalog Attributes: Identifies the information required to complete a service request. This action queries the service configuration to prompt users for details, such as model types and business justifications, to ensure all required fields are populated before submission.
Agentforce IT Service | Get User Profile Details: Retrieves employee profile details from the Employee Record based on the user context. The action returns authoritative information such as name, department or team, manager name, and location to support workflows that require user verification before proceeding with a task.
Agentforce IT Service | Identify Meeting Attendees: Identifies the correct email addresses of the requested attendees based on the user's input, such as attendee names or email addresses. The action resolves the right contacts to use when scheduling a meeting.
Agentforce IT Service | Identify Service Catalog Item: Identifies and selects the correct items from the service catalog by using descriptions. This action searches the catalog based on user input to find matching hardware or services and simplify the request process.
Agentforce IT Service | Post Swarming Summary to Feed Item: Posts the swarming summary of a Slack channel as a new feed item to the Chatter feed for a target record such as an incident or a problem. Key information from other channels is documented in the context of a record, becomes easily accessible to stakeholders, and can be used as the grounding data for other agent actions.
Agentforce IT Service | Propose Resolution Summary for Problem: Proposes a resolution summary for a problem based on related records, feed and swarm data, knowledge articles, and activities. The action helps IT teams to resolve problems more efficiently by providing a comprehensive and data-driven understanding of the issue.
Agentforce IT Service | Propose Resolution Summary for Incident: Proposes a resolution summary for an incident based on related records, feed and swarm data, knowledge articles, and activities. The action helps IT teams to resolve incidents more efficiently by providing a comprehensive and data-driven understanding of the issue.
Agentforce IT Service | Reopen a Ticket: Reopens resolved IT tickets through a conversational interface, eliminating manual navigation by enabling users to request status changes through Agentforce.
Agentforce IT Service | Retrieve Assigned Assets: Identifies and associates hardware with an incident automatically when an employee reports an issue. This action writes a real-time lookup of assigned assets to ensure that tickets include accurate configuration data and accelerate the resolution process.
Agentforce IT Service | Review Approval Work Item: Reviews an approval work item based on the user's decision and optional comments. The action processes the requested approval or rejection and records the user's comments in the work item history.
Agentforce IT Service | Schedule Outlook Meeting: Schedules a meeting on the Outlook calendar based on the user's input. The action captures the required meeting details and creates the calendar event for the specified attendees.
Agentforce IT Service | Search Configuration Items: Searches for configuration items (CIs) in the CMDB based on the user's filter criteria. The action extracts filter values into separate parameters to return the most relevant list of matching CIs and generates a summary of the results.
Agentforce IT Service | Summarize Approval For Employee: Generates a summary of an approval work item based on the available request context. The action helps the reviewer quickly understand the intent, key details, and relevant information needed to make an approval decision.
Agentforce IT Service | Summarize Change Request: Generates a summary for a Change Request record with details such as the description, category, and subcategory. It also includes the risk score, risk level, related records, feed, knowledge, and tasks. This action is triggered when a user requests a summary, to understand the risk and impact of deploying a change request.
Agentforce IT Service | Summarize Incident: Generates a summary for an Incident record with details such as description, category, subcategory, SLA, major incident details, related records, Chatter feed, knowledge, and swarm data. Incident managers or fulfillers can quickly find what they need without parsing the record.
Agentforce IT Service | Summarize Multiple Incidents: Generates a summary for multiple incidents based on criteria such as the status of incidents, and a specific time period in the past from the current date. Use this action to track common patterns, issues, or trends, and analyze incidents effectively. For example, summarize all open incidents in the last two weeks.
Agentforce IT Service | Summarize Open Approvals For Employee: Summarizes all open approval work items for an employee based on the available approval context. The action generates a concise summary of each open approval to help the employee understand the context and take the appropriate action.
Agentforce IT Service | Summarize Problem: Generates a summary for a Problem record with details such as description, category, subcategory, root cause analysis, related records, Chatter feed, knowledge, and swarm data. Problem managers or fulfillers can quickly find what they need without parsing the record.
Agentforce IT Service | Summarize Releae: Summarizes key details of a release record such as description, status, type, and related records, knowledge, and tasks when a user requests an overview.
Agentforce IT Service | Summarize Incident for Employees: Summarizes incidents for employees to help them track the status of the issues they raised. This action improves the employee experience and reduces follow-up calls to the service desk.
Agentforce IT Service | Update Configuration Item: Updates the attributes of a specific configuration item (CI) in the CMDB based on a CI ID or CI name. The action accepts a JSON map of field names and values to apply the requested changes. Use this action when users request to update, modify, or change attributes of a CI.
Unified Catalog | Launch Service Process: Launches a single, URL-based, guided digital intake form for a service process. By capturing all critical information from the user upfront, this form helps agents to immediately create a case or incident without delays, errors, or follow-up.
Sales | Enrich Account in ZoomInfo: Enriches account data from ZoomInfo including employee count, funding information, technologies used, and other data.
Life Sciences | Add Site and Investigator to Study: The Add Site and Investigator to Study action creates a care program site record from a healthcare facility, healthcare provider, and research study record.
Life Sciences | Add Sites Search Results to Study: The Add Sites Search Results to Study action calls the AddSitesToStudy flow to create care program site records from the care site investigator searchable field records and the research study records.
Life Sciences | Generate Healthcare Provider Summary: The Generate Healthcare Provider Summary action provides contextual summaries of the latest changes, interactions, and insights related to healthcare provider accounts.
Life Sciences | Go to Care Program Sites: The Go to Care Program Sites action is a follow-up action to navigate to the care program sites after they're added to a research study.
Life Sciences | Send Site Feasibility Assessments: The Send Site Feasibility Assessments action sends the feasibility assessments to the sites using careProgramSiteIds, omniscriptId, and expiryDate as the input.
Life Sciences | Summarize Site: The Summarize Site action summarizes the details of a healthcare facility, such as its location, facilities, specialties, resources, accreditations, performance, and quality compliance in the context of a research study.
Media | Create Media Opportunity: Creates an opportunity and an ad opportunity for a media proposal or an advertisement campaign. The opportunity is associated with the account of the advertiser who the media proposal or advertisement campaign is for. The ad opportunity is associated with the account of the agency that represents the advertiser.
Media | Create Media Proposal: Creates a quote and an ad quote for a media proposal, and adds media products for the proposal. The proposal is associated with the opportunity of the advertisement campaign that the proposal is for. The proposal includes key details such as budget, start and end dates, and price book. For proposal products, details such as price book, quantity, and start and end dates are added.
Media | Get Media Products: Gets a filtered list of media products that are part of a price book and match a search phrase. Users can provide various search phrase combinations that can include product name, SKU, code, and price range. If users aren't sure about the product they are looking for, the action can also get products based on price book name.
Media | Get Media Products Based on Targeting: Gets a list of media products that are part of a price book and match a targeting criteria. The targeting criteria can include a list of targeting values. If users aren't sure about the products they are looking for, they can expand their search criteria by providing a list of targeting categories and segments.
Media | Media Opportunity Summary: Summarizes media opportunities with actionable information such as closure risks, advertiser or agency sentiment, current stage, and next best steps.
Manufacturing | Check Coverage for Parts: Validates and retrieves serializable parts within an asset's hierarchy. If none are found, retrieves the non-serializable parts with matching names covered by the main or the root asset's warranty terms and checks the coverage of the same.
Manufacturing | Create Asset Appraisal: Creates an appraisal record for an asset and an associated appraisal item based on the customer's selected asset that they want to trade-in and opportunity context. The action captures the opportunity ID, the specific record ID of the customer's trade-in asset, and the asset's condition status.
Manufacturing | Create Opportunity with Products and Preferred Seller: Creates an opportunity, associated opportunity products, and opportunity preferred seller records for a customer based on inputs that define the core record identifiers and data collections for the flow, including the account, product, and preferred dealer IDs, alongside an Apex collection containing the pricebook mapping details required for opportunity creation. Captures the details of the newly generated records, including the opportunity's core attributes, the associated preferred seller record, and specific line-item data such as the opportunity ID, price book entry, product ID, quantity, and unit price.
Manufacturing | Create Product Request: Create product requests for specified products and quantities from the given location to address shortage scenarios. This action supports the creation of product requests to streamline inventory management.
Manufacturing | Create Quote from Opportunity: Creates quotes and quote line items for an opportunity using its ID as an input and its associated opportunity products with a manual discount percentage.
Manufacturing | Create Service Quote: Creates a comprehensive quote for specified products by evaluating requested quantities, selected parts, manual discounts, and applicable coverages for each item.
Manufacturing | Get Pricebooks From WorkType: Gets the product IDs associated with a worktype and subsequently gets the pricebook records containing all the retrieved product records.
Manufacturing | Search Product And Parts: Gets products and associated parts or accessories based on inputs that define the search parameters, including the text-based query string or the base product ID, while using boolean flags to distinguish between text searches and product-related component searches. Provides a list of matching products with their respective pricing, or prompts users for more specific search queries.
Manufacturing | Get Warranty Claims for Account: Gets a list of warranty claims associated with a specific account, using optional parameters such as an asset ID, an asset serial number, or a claim number.
Manufacturing | Search Dealers for Products: Gets the dealer product searchable field records based on location and user specified search criteria and filters the records to find dealers that offer user requested products in a service territory. Requires City and Postal Code inputs from the user utterance to filter the Dealer Product Searchable Field records.
Manufacturing | Search Customer Accounts: Validates the agent session and searches for customer accounts by using a provided email address, phone number, or account name.
Manufacturing | Summarize Inventory Details: Summarizes the available inventory for an item at a specific location, including details such as quantity and stock status. The action uses inputs such as product name, inventory location name, and stock status, where stock status values can be shortage, excess, or optimal. Users can provide one or a combination of these inputs to refine the summary.
Manufacturing | Summarize Sales Agreement Performance: Generates a summary of the performance of a sales agreement, including the deviations between the planned and actual quantity and revenue values in the sales agreement.
Marketing | Create Journey Content: Creates marketing content that's tailored for each customer in a journey using the Journey Decisioning agent template.
Marketing | Select Journey: Chooses the optimal journey for each subscriber to maximize message engagement.
Marketing | Get Marketing Object Metadata and Planning Hierarchy: Looks up the field-level setup of marketing planning objects and the available planning hierarchies before the agent creates records. It returns the fields that a user can set on a program or goal. It also returns the hierarchy groups, levels, and per-level create access for that user. This action is read-only.
Marketing | Manage Marketing Planning Records: Creates marketing programs and goals and places them in a planning hierarchy. It creates each record at the hierarchy group and level that the user names. It links new records to the parent programs or goals in the request, and confirms each record as it goes.
Net Zero Cloud | Combine Snippets: Combines two environmental, social, and governance (ESG) reports related snippets to generate the final snippet content. One snippet is from the document clause record and the other snippet is from the agent conversation.
Net Zero Cloud | Create Sustainability Task Group: Create a sustainability task group and notify sustainability stakeholders about environmental, social, and governance (ESG) data collection activities.
Net Zero Cloud | Draft Disclosure Responses: Answers queries and drafts snippets related to environmental, social, and governance (ESG) topics. These include the organization's policies and practices regarding climate change, energy efficiency, greenhouse gas removal, renewable energy, and other sustainability issues.
Net Zero Cloud | Draft Snippet For Question: Answers queries that originate from assessment question records directly associated to a document clause record. These include providing information relevant to environmental, social, and governance (ESG) reporting.
Net Zero Cloud | Manage Sustainability Task Group: Manage a sustainability task group by sending follow-up emails to stakeholders with incomplete tasks in an existing sustainability task group. Improve stakeholder engagement and completion rates for environmental, social, and governance (ESG) data collection.
Omnistudio | Diagnose Components (Pilot): Diagnoses Omnistudio components, namely, Flexcards, Omniscripts, Integration Procedures, and Data Mappers. Identifies potential warnings and errors that may occur at runtime and provides information about how to fix them.
Omnistudio | Get Org Objects (Pilot): Returns a list of sObjects associated with the given use case and, if needed, suggests new objects users must create to fulfill a requirement.
Program and Case Management | Create Participant Benefits: Generate a benefit disbursement record that includes all required records for program enrollment and benefit assignment. If any related records are missing, the flow creates them as needed.
Program and Case Management | Prepare Participant Notes: Fixes spelling and grammatical errors, applies formatting and best practices from the provided template, and structures the information to highlight key points, decisions, and action items in the notes of a meeting with a participant or client.
Public Sector | Automatically Resolve Complaint: If the incident reported in a complaint occurred more than two years ago, asks whether the user wants to resolve the complaint. If the user confirms, sets the complaint status to Resolved.
Public Sector | Check For Regulatory Code Violation: Evaluates user input against regulatory code documents to determine violations. Cross-references the input with relevant code sections and provides a clear, concise determination, including specific clause references and supporting explanations.
Public Sector | Create Assessment Question Response: Creates an assessment question response record for an assessment question by using the response text, and relates the assessment question response to the assessment.
Public Sector | Create Candidate Leads: Creates lead records for potential candidates for a vacancy. In the lead records, includes information from the party profile of applicants who submitted the application forms selected by the user.
Public Sector | Draft Visit Summary Email: Generates a professional email summarizing the visit with details about the visit, regulatory code violations, enforcement actions, and regulatory fees along with their due dates and resolution notes.
Public Sector | Email Candidate Leads: Sends emails to candidates who are identified as leads for a vacancy to invite them to check the recruitment posting and apply for the job.
Public Sector | Email Complaint Participant: When the user asks to notify the complainant or the perpetrator, generate an email based on the given template. Show the body of the generated email to the user and get approval. Send the email to the complainant or perpetrator.
Public Sector | Get Examination and Competency Details for Skills: Gets a list of IDs for examinations and competencies that match the input skill names. Then, sends those details to the associated prompt template. Outputs a JSON list of skill-to-record ID pairings.
Public Sector | Get Regulatory Code Violation Details: Analyzes the user's request to identify the relevant section or clause from the regulatory code violation documents. Extracts and provides detailed information about the identified section or clause, ensuring the response directly addresses the user's question and adheres strictly to the regulations outlined in the documents.
Public Sector | Preview Hiring Managers: Previews the hiring managers that are associated with the application form. Before sharing an application form with the hiring managers, always call this action to preview hiring managers who are also application form participants.
Public Sector | Share Application Form with Hiring Manager: Shares an application form with employees having a participant role as Hiring Manager. Gets recruitment requisition participants from the related recruitment posting. Based on the recruitment requisition participants, creates application form participant records.
Sales | Qualify Account: Qualifies an account based on research data and qualification criteria. Analyzes account signals internally to generate an Account Point of View summary that explains whether the account qualifies for prospecting.
Marketing | Refine Campaign Preview: From user input, refines and replaces campaign preview steps associated with a brief, such as message channels and content, before the campaign is saved.
Security | Retrieve Detected Findings: Retrieves detected findings based on the start date and end date. Optionally, filters by userId or category. Returns all findings that match the time frame.
Security | Retrieve Investigation: Returns a complete security investigation, including findings, timeline entries, and impacted instances.
Security | Retrieve Investigations: Returns a list of security investigations, with optional filters for stage, category, or date range. The information can include associated findings, timeline entries, and remediation plans.
Sales | Return Calendar Link: Adds a calendar link to the agent’s response, allowing web prospects quickly schedule a meeting that is convenient for them.
Revenue | Create Amendment Quote: Creates an amendment quote in Revenue Cloud and then generates a related amendment opportunity for an account with active assets. An amendment quote, also known as an add-on quote or a change quote, increases the quantity of assets a customer has purchased or modifies some aspect of those assets.
Revenue | Create Initial Quote: Creates a Revenue Cloud quote for a new business from an opportunity. This quote differs from an amendment quote or a renewal quote that modifies or extends an existing business. Sales reps can create multiple quotes for an opportunity.
Revenue | Explain Invoice Line: Provides a detailed, easy-to-understand explanation of a specific invoice line, including the purpose of the charge, the billing reason, and the calculation method.
Revenue | Get Account Billing Summary: The Get Account Billing Summary agent action generates a comprehensive financial health overview for a specific account. This action identifies active disputes and collection plans while providing insights into payment history, expiring payment methods, and aging metrics to help prioritize collection efforts.
Revenue | Get Dunning Strategy: The Get Dunning Strategy action assesses an account’s financial health, ongoing collection efforts, prior interactions, payment history, and open disputes to recommend the most effective engagement and recovery approach. Using AI-driven insights such as invoice risk scores and communication patterns, it suggests the optimal dunning orchestration, including targeted reminders, calls, or, in higher-risk scenarios, actions like applying late fees, suspending services, or escalating the case.
Revenue | Create Renewal Quote: Use the Create Renewal Quote action to create a renewal quote for active and renewable assets in Revenue Cloud.
Revenue | Get Account Assets: Retrieves the active and term-defined renewable assets associated with an account. Active assets are assets with one-time or evergreen product selling models or assets with term-defined product selling models that haven’t reached their lifecycle end date. Renewable assets are assets with term-defined product selling models.
Revenue | Get Usage Details: Gets the details of the consumption overages for up to five usage resources associated with an account by using the account ID.
Revenue | Find Products: The Find Products agent action searches the product catalog for items that match a user’s input query. It returns product names, IDs, and structured details to support conversational product discovery in Agentforce.
Revenue | Get Account Balance: The Get Account Balance agent action retrieves a customer’s billing balance for a specific account by summarizing posted invoices, credits, debits, payments, and refunds so that users can quickly understand the net amount due in a conversational billing inquiry.
Revenue | Get Invoice Document PDF: The Get Invoice Document PDF agent action retrieves the most recent posted invoice for an account and provides access to its associated document. Users can view or download the latest invoice PDF when it's available.
Revenue | Retrieve Next Due Payment: The Retrieve Next Due Payment agent action identifies the next upcoming payment for an account by returning the earliest due date and the total amount due so that users can quickly understand what a customer needs to pay next.
Revenue | Generate Payment Plan Details: The Generate Payment Plan Details agent action returns the payment installment schedule for an invoice so that users can understand how an invoice is paid over time, including installment amounts, dates, and statuses.
Revenue | Get Product Details: The Get Product Details agent action retrieves structured details for a list of product IDs so that users can refine results and apply customer-specific filters during a Product Selection conversation in Agentforce.
Revenue | Get Rule-Based Product Recommendations (Beta): The Get Rule-Based Product Recommendations agent action returns recommended products for a quote based on configuration rules in Product Catalog Management (PCM). The action evaluates the products already added to a quote, applies relevant recommendation rules, and returns a list of recommended product IDs and names that can be used in Product Selection workflows in Agentforce.
Revenue | Query Contract Document: Finds and retrieves relevant information from a contract document based on the user's question. Answers questions related to contract documents, including terms, obligations, and clauses.
Revenue | Recall Approval Submission: Recalls a submitted approval request, withdrawing it before a final decision. Use this when an in-progress approval needs to be canceled, such as when it was submitted in error, the record changed, or approval is no longer required. The user provides the approval submission name and may optionally add recall comments. The action returns a result summary and success indicator confirming whether the recall was completed.
Revenue | Summarize Contract Document: The Summarize Contract Document agent action calls Summarize Contract prompt templates of type Document Summary to create clear and business-focused contract overviews.
Security | Save Investigation: Creates a Security Agent investigation record, or updates an existing investigation with investigation details, category, timestamps, and risk assessment.
Security | Save Remediation Plan: Creates a Security Agent investigation remediation plan record or updates an existing one with remediation plan details, associated items, and active status.
Sales | Search Conversations: Searches voice and video call transcripts for specific topics or keywords. Agentforce Sales Management calls this action.
Slack | Start a Conversation with an Agent in Slack: Starts a conversation with an Agentforce agent in a Slack direct message. After the conversation is initiated, the recipient can interact with the agent.
Slack | Summarize a Channel: Provides a summary of channel messages for a specified channel and time frame. Channel summaries default to seven days and can be up to 30 days.
Automotive | Summarize Customer Account: Summarizes an account for automotive industry use cases, with details such as the account’s assets, vehicles, driver performance, households, engagements, and interests.
Automotive | Summarize Financial Account for Customer: Summarizes the financial account for automotive industry use cases, with details such as the financial account’s parties, statements, fees, balances, transactions, and vehicles.
Automotive | Summarize Customer Finance Profile: Summarizes the financial information of an account, with details such as the account’s financial accounts, application forms, cases, and opportunities.
Inventory Management | Summarize Inventory Details: Summarizes the available inventory for an item at a specific location, including details such as quantity and stock status. The action uses inputs such as product name, inventory location name, and stock status, where stock status values can be shortage, excess, or optimal. Users can provide one or a combination of these inputs to refine the summary.
Sales Agreement | Summarize Sales Agreement Performance: Generates a summary of the performance of a sales agreement, including the deviations between the planned and actual quantity and revenue values in the sales agreement.
Swarming | Begin Swarm: Initiates a swarm in response to a swarm request. Used to get subject matter experts collaborating to respond and resolve customer issues.
Swarming | Get Swarm Type and Slack Access Details: Gets the details of the Slack workspace, whether the swarm is in a dedicated or new channel, and returns the appropriate Swarm Channel type. Then, using the Slack workspace ID, confirms the running user's access to create a Salesforce channel.
Swarming | Reopen Swarm Procedure: Reopens a swarm, updating member statuses and handling Omni Skills routing as needed. This action supports multiple reopen strategies and ensures there are no duplicate member assignments.
Swarming | Send Slack Udpates: Sends updates to Slack channels and members as part of the swarming process. Handles member invitations, message updates, and error reporting for Slack integration.
Swarming | Unarchive Channel: Unarchives the Slack channel associated with a Swarm, ensures that the channel is eligible, updates related records and members, checks Slack integration, retrieves channel and swarm details, and handles errors.
Timesheets | Get Timesheet Details: Gets the details including the timesheet, timesheet entries, and timesheet validation errors for a date provided by the user.
Timesheets | Verify Or Submit Timesheet: Initiates the process of verification or submission of a timesheet. The response indicates whether the timesheet has been sent for verification or submission.
Sales | Update AI-Generated Action Item: Updates the status of an AI-generated action item record. The action tracks the lifecycle and completion state of recommendations and tasks that Agentforce Sales agent generates. Agentforce acts on these recommendations automatically, or human users review them, depending on org settings. Agentforce Sales Management calls this action.
Security | Update Remediation Plan Item: Updates an existing Security Agent investigation remediation plan item with new values for status, execution details, and other modifiable fields that use the remediation plan item ID.
Sales | Validate Prospects: Validates a prospect against Ideal Customer Profile criteria by using LLM-based evaluation. Returns whether the prospect is valid, a confidence score, and an evaluation summary.
Sales | Verify Sales Customer: Compares the customer’s entered code with the generated verification code from the Send Verification Code agent action to determine if the customer is verified.
Volunteer Management | Find Unfilled Shifts: Identifies volunteer initiatives and job positions that have unfilled shifts. Return unfilled shifts within a date range and a capacity threshold.
Volunteer Management | Match Volunteers to Shifts: Find suitable volunteers to fill open shifts based on the time, location, and qualification requirements. Return volunteers that meet the shift requirements and generate a list from the results to use as a contact list to request that the volunteers sign up for the open shift.
Types of Voice Call Records: For each incoming call, you can have these types of related voice call records.
Set Up Connections in the Legacy Builder: To get started with deploying your agent to channels, set up connections. Each connection includes behind-the-scenes instructions that help an agent reason and respond for one or more channels. Each connection contains adaptive response formats that help your agent structure responses and deliver multimedia content, such as images, buttons, links, and videos. You can find adaptive response formats and other settings, including the Omni-Channel flows that route conversations to and from an agent, in your connection settings.
Example: Send a Rich Choice Response to a Customer With a Flow: In this example, you send product options using an autolaunched flow and rich choice response. You also help the agent decide whether to present text-based product options as buttons or a list.
Standard Adaptive Response Format Reference: Standard adaptive response formats are included with standard connections and are built to work with specific channels, interfaces, or apps.
Adaptive Response Format: Rich Choice Response: A rich choice response is a question with a list of records specific to the end-user. Currently, rich choice response is available for Agentforce Service Agent only.
Adaptive Response Format: Rich Link Response: A rich link response is a web page link with a title and image. Currently, rich link response is available for Agentforce Service Agent only.
Adaptive Response Format: Time Picker: A time picker response is a list of available time slots. Currently, time picker response is available for Agentforce Service Agent only.
Agent Script: Get to know the language for building agents in Agentforce Builder. Whether you’re working in Canvas view, Script view, or chatting with Agentforce to build and customize your agent, you’re writing Agent Script. Use Agent Script to build predictable, context-aware agent workflows that don’t rely solely on interpretation by an LLM.
Search the Web with Agentforce: Use the Search the Web action to retrieve real-time information from the internet for your agents and prompt templates. Search the Web lets you ground agents with current data like news, market trends, and competitor details that aren't available in Salesforce.
Select the Right Persona for Your Agent: Selecting the right persona for your agent shapes how every call with your customers goes, so it's worth selecting deliberately rather than by first impression.
Create an Agent: Create intelligent, trusted, and customizable agents for your customers and employees.
Upgrade an Agent from the Legacy Builder to the New Builder: Create a new version of an agent from the legacy Agentforce Builder in Setup in the new Agentforce Builder in Agentforce Studio, with just a few clicks. The new version contains all of the original agent's subagents, actions, system messages and settings, data, and connections, converted to Agent Script. Upgrading an agent doesn't affect the original agent in the legacy builder. You can continue to access and edit the agent in the legacy builder.
Troubleshoot Agent Upgrade Errors: When you upgrade an existing agent, Agentforce does most of the work for you. But metadata and syntax differences between the new and legacy Agentforce Builders can still cause validation errors when you save or commit your upgraded agent. Learn more about common issues and how to solve them.
Select Agentforce Model Option: We recommend using the Salesforce Default model option for Agentforce, which enables Salesforce to select the best model mix for agent accuracy and performance. But you can choose an AWS-Hosted or Google Gemini model option instead.
Simulate and Live Test Modes: Agentforce Builder includes two different modes to test your agent: Simulate and Live Test. Simulate mode helps you verify that your agent has the right configuration without risking any changes to your actual data or org. Live Test mode actually runs the actions and changes your org so you can see how the agent performs when it’s active.
Include Custom SIP Headers and REFER Support: Increase the amount of information passed with SIP requests with custom SIP X-headers and handle disconnects more gracefully with SIP REFER.
Select Your Test Data: Evaluate your agent's performance by using AI-generated test cases or uploading your own test cases in a CSV template. AI-generated test cases offer two data options: generating tests based on subagents and actions, or generating tests based on the knowledge available to the agent via an Agentforce Data Library.
Create a Test Suite: A test suite is the starting point for evaluating your agent. Name your test suite, select the agent to test, and choose whether you're evaluating individual responses or full end-to-end conversations.
Create Custom Scorers: Custom scorers evaluate AI agent sessions against specific criteria that you define, like tone, brand voice, or business outcomes. Use them alongside default scorers to get a complete picture of agent performance.
Default Scorers: Default scorers measure your AI agent's performance across response accuracy, subagent assertion, and action assertion.
Select Scorers: Scorers measure your AI agent’s performance across key areas. Default scorers are always tested, but you can select quality metrics to focus your tests on specific areas.
Generate Tests from Knowledge Sources: When your agent is connected to an Agentforce Data Library, you can generate test scenarios directly from the content linked to that library. You can base test cases on multiple data sources including articles and knowledge fields or uploaded files.
Test Prompt Templates: Evaluate your prompt templates at scale with batch testing in Agentforce Studio Testing Center. Upload test data, select scorers, and run evaluations across many inputs at once.
Create a Prompt Template Test Suite: Start a batch test by naming your test suite and selecting the prompt template and version you want to evaluate.
Add Prompt Template Test Data: Provide the test inputs your prompt template is evaluated against. Download a CSV template preconfigured for your prompt template's input variables, fill in your test cases, and upload it to run your evaluations.
Select Prompt Template Scorers: Choose the scorers you want to apply to your test cases. Each scorer evaluates a different dimension of your prompt template's output and returns a pass or fail result.
Run Quality Scorers: Response quality scorers measure your agent's accuracy and relevance.
Set Up and Run Tests in Agentforce Studio (Beta): Use Testing Center in Agentforce Studio (Beta) to test AI agents or prompt templates at scale. Test agents by simulating conversations across many scenarios at once. Or, test prompt templates by running batch evaluations across many inputs. Apply scorers to quickly evaluate responses and identify quality issues.
Test Agents: Test your AI agents at scale with a flexible, AI-driven testing workflow in Agentforce Studio Testing Center. Build test suites, generate test cases, and run evaluations across many scenarios at once in a spreadsheet-inspired interface designed to help you validate and debug agent conversations.
Define Test Conditions: Choose how your agent accesses data during testing and configure optional conditions like conversation history and context variables to simulate more realistic interactions.
Generate Tests from Subagents and Actions: Evaluate how well your agent handles key subagents and actions. Generate relevant, test-ready utterances by using your subagent and action metadata as well as information from connected data libraries.
Edit a Subagent in the Asset Library: Edit a subagent directly in the Asset Library. The edits don't modify versions of the subagent already in use, as they're copies of the original subagent.
Choosing an AI Model for a Subagent: Choose the right AI model for each part of your agent to balance response speed, accuracy, and reasoning depth.
Test Your Agent in Agentforce Builder: Manually testing your agent in Agentforce Builder provides a detailed, step-by-step trace of your agent's interactions and processes. Use the Preview panel to chat with your agent, then review subagent selection, action execution, and reasoning on the Interaction Summary panel. For more details about any step, converse with the Agentforce Assistant.
Build an Agent Testing and Observability Plan: A test plan and a monitoring plan turn your definition of agent success into something measurable. The test plan helps identify regressions before they reach production and shows that the agent is ready for deployment. The monitoring plan helps you recognize and address areas for improvement after go-live.
Agent Testing Strategy: Define what your agent is built to achieve, map the observability tools that measure it, and build the plan that connects your KPIs to testing and monitoring before your agent goes live.
Agentforce Testing Center: Build confidence in your AI agents and deliver seamless, effective interactions with thorough testing. Easily generate custom test scenarios and leverage our expert evaluation tools to assess response accuracy, conversation quality, subagent recognition, action execution, and knowledge retrieval. Identify issues early and fine-tune your agent’s behavior with precision.
Add Test Data: Add test cases to evaluate your agent’s performance by either generating them with AI or uploading your own scenarios. AI-generated test cases offer two data options: generating based on subagents and actions, or generating from knowledge available to the agent via the Agentforce Data Library.
Considerations for Testing Center: To use Testing Center, consider supported functionality, usage, limitations and allowances, limits, and other issues.
Create a Test: Fast-track your testing with AI generated testing scenarios or create and upload your own. Test your agents against subagents and actions or data available in Agentforce Data Library.
Choose Evaluations: Evaluations measure your AI agent’s performance across key areas. Default evaluations include response accuracy, subagent assertion, and action assertion. To better match your testing goals, you can add quality-focused evaluations like completeness, coherence, conciseness, latency, and instruction adherence. Selecting the right mix of evaluations gives you the insights you need to strategically refine and improve your agent.
Generate Tests from Knowledge Sources: When your agent is connected to an Agentforce Data Library, you can generate test scenarios directly from the content linked to that library including articles and knowledge fields or uploaded files. Testing center generates up to 100 question-and-answer style test cases based on the knowledge content available to the agent.
Set Up and Run Tests in Setup (Legacy): Efficiently validate your agent’s performance at scale, whether you want to fast-track testing with AI-generated scenarios or manually craft and upload your own. You can generate realistic test cases for specific subagents and actions, or data linked through the Agentforce Data Library.
Define Test Conditions: When you upload test cases from the CSV template, defining test conditions such as conversation history and context variables helps provide additional context. These variables help simulate more realistic user interactions, making your tests more reliable and reflective of actual agent behavior.
Generate Tests from Subagents and Actions: Evaluate how well your agent handles key subagents and actions before launch. Generate relevant, test-ready utterances using your subagent and action metadata as well as information from connected data libraries. Simply choose the subagents you want to test, define the test focus with a brief description, and let the system generate up to 100 test cases.
Use Test Results: After a test run completes, review evaluation scores to identify where your agent or prompt template is falling short. Understand what each scorer measures, how scores are calculated, and what to do when results are low .
Subagents: A subagent is a job an agent can do. Learn about the parts of a subagent and how to use them to define an agent’s logic, reasoning, and conversational capabilities.
Add a Subagent to Your Agent from the Asset Library: Add a standard subagent to your agent to start from an example built to handle a common business use case. Or add a custom subagent that you built yourself. When you add a subagent to your agent from the asset library, a copy of the subagent is added to your agent.
Create a Custom Subagent: You can create a custom subagent from the asset library or from within an agent in Agentforce Builder. When you create a subagent from the asset library, it’s available to add to multiple agents and versions. When you create a subagent from within an agent in Agentforce Builder, it’s available only to that agent.
Delete a Subagent from the Asset Library: Delete a custom subagent from the asset library when you no longer want it to be available to add to your agents. You can’t delete a standard subagent from the asset library.
Manage Subagents: Add and remove subagents from your agent. Create custom subagents. Manage subagents in the asset library.
Subagents: A subagent is a particular job an agent can do. Collectively, the subagents assigned to your agent define the capabilities your agent can handle. Learn about subagents in Agentforce Builder and the legacy builder.
Remove a Subagent from Your Agent: Make sure that your agent continues to deliver helpful and relevant responses by removing subagents you no longer need.
Subagent Classification and Routing: Learn more about the Agent Router and how to customize it to control how your agent selects subagents in a conversation.
Data Masking Limitations in Agentforce: In Einstein Trust Layer, pattern-based and field-based data masking for large language models (LLMs) is disabled for agents.
Configure Service Agent Access: Learn how Agentforce Service agents control data access. Set up an agent user for your Agentforce Service agent and assign permissions, so your agent has everything it needs to do its job.
Enable User Profile: Enable user profile context in Agent Script so your agent can personalize responses using available profile information about the users it interacts with. The profile information that's available depends on the type of agent and the user record associated with the interaction.
Variables in the Legacy Builder: To improve security and ensure consistent agent behavior, use variables in filters, instructions, and as input for actions. You can use variables to control how your agent uses subagents and actions and to provide more information about a conversation.
Create a Custom Variable for Your Agent in the Legacy Builder: Control how an agent uses subagents and actions by mapping an action output to a custom variable. For example, when you assign a variable to an action input, your agent can use only the assigned variable to fill the value of the input.
Standard Variable Reference for the Legacy Builder: Some agents include pre-built variables, called standard variables. Explore your agent's standard variables on the Context tab in the legacy Agentforce Builder.
Manage Agent Versions in the Legacy Builder: Create multiple versions of an agent to make and test changes without taking the active version out of production. Or create a version to save as a backup, so you can quickly revert to a previous version. When you create a new version of an agent, Salesforce creates independent copies of the agent's subagents, actions, and other assets. Every agent and agent version has its own assets, and assets aren't shared.
Commit an Agent Version: Commit an agent version to save a snapshot of your agent and prepare your agent for use outside of Agentforce Builder. Your agent version must be committed to activate it.
Compare Agent Versions: Compare multiple versions of an agent and copy improvements from one agent version to another. You can accept or decline changes individually or all at once.
Delete an Agent or Agent Version: Delete an agent or agent version to remove it entirely from Agentforce Builder. You can’t recover the agent after its deleted.
Create a New Version of an Agent: Create multiple versions of an agent to seamlessly iterate on your agent’s design. Create a version to experiment with different instructions, subagents, or actions and preserve your current configuration. Or create a version to save as a backup, so you can revert quickly.
Versioning and Editing Agents: Learn about the different states of agents as you create and iterate on your agents, including editable states and how to version your agent.
Create the Channel for Agentforce Voice with PSTN: Create a Messaging channel for incoming calls that need to be transferred to the Agentforce agent. In the channel, map the procured PSTN phone number to the inbound Omni-Channel flow. When an incoming call is transferred using this number, Omni-Channel uses the Omni-Channel flow to transfer the call to the agent assigned to the channel.
Create the Channel for Agentforce Voice with SIP or Dynamic Routing: Create a Messaging channel to manage incoming calls transferred to the Agentforce agent. To route inbound calls to the agent using the SIP address or Dynamic Routing number, create a Messaging channel that maps the SIP address or Dynamic Routing number to the Omni-Channel flow. When an incoming call is transferred using this address or number, the Omni-Channel flow routes the call to the agent assigned to the channel.
Define Agent Success: Agentforce surfaces a lot of data. Without a clear sense of what your agent is supposed to accomplish, that data creates noise instead of direction. Defining success before you start measuring is how you know which signals matter, which ones to ignore, and when your agent is actually improving.
Apply Policies by Agent: Target which agents a policy applies to for API connections and MCP server connections. Without this configuration, a policy that protects a connection applies to all agents that call it.
Create an Agentforce Gateway Policy: Create a policy from a prebuilt template and follow a few simple steps to build a new policy that fits your governance needs. You can apply the policy immediately or save it and apply it later.
Default Policy Execution Order for Agentforce Gateway: The default policy execution order makes sure that you protect your connections as intended. The execution order is based on the policy type, application method, and timestamp.
Delete an Agentforce Gateway Policy: If you no longer need a policy, you can delete it. Deleting a policy removes protection from all associated connections and agents.
Manage Agentforce Gateway Policies: Understand the default policy execution order and learn how to manage existing policies. You can edit a policy’s details or adjust the connections it protects as well as the agents it targets.
Agentforce Gateway: Use Agentforce gateway policies to protect connections to your external APIs and Model Context Protocol (MCP) servers. Control access, set usage limits, and ensure compliance.
Agentforce Gateway Policy Templates: To create a policy, start from a template. Available templates depend on whether you protect API or MCP server connections.
Agentforce Grid: Agentforce Grid is a powerful spreadsheet-like tool for orchestrating and testing AI features, such as prompts and agents, quickly within Salesforce. You create AI workflows where each column builds on the last, taking you from raw data to insight to action.
Use Grid Agent to Create or Analyze a Column: Grid Agent is an AI agent in the Agentforce Grid sidebar that creates and analyzes columns from plain-language requests. Describe what you want, review the proposed changes, and approve them. Grid Agent handles the configuration.
Agentforce Grid Columns: Agentforce Grid uses data columns to connect your CRM and Data 360 data and action columns to do things with the data. The columns build on each other, from left to right.
Create an Agentforce Grid Worksheet: After you have Einstein enabled and the permission assigned to your users, you’re ready to get started exploring solutions.
Send List View Records to an Agentforce Grid: Send records from Salesforce list views directly into Grid worksheets so you can quickly run AI workflows, formulas, and actions on your data.
Manage an Agentforce Grid Template: When you update a Grid that's published as a template, you can push those changes to the template without creating a duplicate. Updates don't affect grids previously created from the template. You can also delete templates that aren’t in use. You can delete templates that you created, but not templates created by others or by Salesforce. Deletion is permanent and there's no way to restore a deleted template. Grids previously created from the template continue to work normally.
Publish an Agentforce Grid as a Template: Save a grid as a template so others in your org can start from your setup without rebuilding columns, prompts, and logic. Templates store column configuration only. They don't include row data, run history, or cell outputs.
Set Up Agentforce Grid: To begin using Agentforce Grid, first enable Einstein and assign the necessary permissions.
Agentforce Grid Templates: Grid templates are reusable starting points that package a fully configured Grid, including columns, prompts, queries, and logic. By using templates, anyone in your org can get started without building from scratch.
Build and Test Observability Tools: Build and Test tools let you observe agent behavior during development, before production traffic exists. Testing in Agentforce Builder enables you to iterate quickly on individual conversations, while Testing Center automates testing at scale.
Agentforce Observability Infrastructure: Before any observability tool can show or process data, the underlying agent telemetry infrastructure must be enabled and collecting activity data in Data 360.
Observe Phase Observability Tools: Once your agent is deployed, observability shifts from controlled evaluation to live monitoring. This phase focuses on identifying and tracking trends across all sessions, investigating specific conversations, and drill into exactly what happened at the step level.
Agentforce and Generative AI Resources: This page provides a curated set of links to videos, Trailhead, Developer Guide, and other resources to help you learn about and implement Agentforce and generative AI within your Salesforce org.
Understand Agentforce Observability KPIs and Tools: Observability isn't a single tool. It's a layered toolchain that includes infrastructure that captures agent activity, development tools to test your agent pre-deployment, and monitoring tools that surface trends and session-level details after deployment.
Agentforce Voice: Agentforce Voice enables Agentforce Service agents to speak and understand voice conversations with customers. It is designed to understand customer intent from spoken input and take actions to resolve customer queries.
Add Agentforce as the Voice Collaborator: Create an orchestration AI agent that hands the voice contact to Agentforce on the first turn, with Agentforce using its own voice. Save the agent configuration and create the agent. Connect streams PCM audio at 8 kHz, 16 kHz, or 24 kHz.
Allow the Application on the Security Profile: Allow the A2A application on the Connect AI agent’s security profile, and associate that profile with the agent. Connect can’t hand the voice contact to Agentforce until the profile allows the application.
How Amazon Connect and Salesforce Work Together with Real-Time Streaming: Agentforce Voice allows a Salesforce AI agent to join live calls in Amazon Connect. Agentforce Voice runs in Salesforce, separate from your Amazon Connect instance. The two services communicate over a secure WebSocket connection using Amazon’s Agent-to-Agent (A2A) protocol.
Associate the Application with the Amazon Connect Instance: Add the registered A2A application to this Amazon Connect instance so the instance can open the Agentforce WebSocket. Use the application ARN returned when you registered the application.
Enable Auto-Hangup for Agentforce Voice: Set up your voice-enabled agent to end a call automatically when it has fully addressed the caller's request. Enable this behavior per agent with Agentforce Script. If you use partner telephony, also turn on a setting that returns the call to your telephony vendor, and prepare your interactive voice response (IVR) flow to read the disconnect reason and route the returned call.
Agentforce Voice Best Practices: Voice-enabled agents require distinct conversational design patterns compared to text-based agents to handle spoken interaction effectively.
Create the Channel for Agentforce Voice with Real-Time Streaming: Create a Messaging channel to manage incoming calls transferred to the Agentforce agent. To route inbound calls to the agent using real-time streaming, create a Messaging channel that maps the incoming phone number to the Omni-Channel flow. The flow specifies the agent to route the incoming call to.
Agentforce Voice Considerations: When creating, testing, and managing voice-enabled agents consider the supported languages, usage and other limits.
Create the Amazon Lex Bot: Create an Amazon Lex V2 bot that sends a voice contact to the Agentforce agent. Keep the bot and the Q in Connect assistant in the same region.
Create the Security Profile: Create a dedicated Amazon Connect security profile for the third-party A2A integration. This profile becomes the authorization boundary between the Amazon Q orchestration agent and the external Salesforce A2A application.
Dynamic Routing in a Partner Contact Center: For Dynamic Routing, your partner contact center routes incoming calls to the Agentforce agent using a dynamic routing phone number via a PSTN call transfer.
Configure Telephony Vendor Settings When Using Dynamic Routing: If supported, configure your telephony system to create a VoiceCall record in Salesforce for each incoming call. Then, obtain a dynamic routing number—a temporary, call-specific phone number—to transfer the call to the Agentforce agent.
Procured Phone Number Statuses: Review the possible statuses of procured phone numbers. To route a call to an agent with a procured phone number, the number must have a Live status.
Configure the Channel and Telephony Settings for PSTN: Create a channel that's used to manage inbound calls routed to the Agentforce agent. If supported, configure your telephony system to create a VoiceCall record in Salesforce for every incoming call, and then transfer each call to the Agentforce agent using the procured phone number. When needed, transfer the call to a human service rep.
Publish the AI Agent Version: Publishing a version applies the new agent configuration to new contacts. Use that same version on the security profile and in the contact flow. A new voice contact that reaches the flow then opens one WebSocket to Agentforce for the session.
Configure Authentication If You Aren’t a Salesforce Voice with Telephony Providers Customer: If you aren’t a Salesforce Voice with Telephony Providers customer, deploy the SalesforceJwtRotatorNonScvStack CloudFormation stack to the same AWS account and region as the instance. Set the instance, org, and channel parameters. After the stack is created, replace the private-key placeholder in the <identity>-salesforce-secret secret with your RSA private key. Then run the rotator and confirm that the JWT secret contains a token before you use its ARN. The rotator generates the token, rotates it every 55 minutes, and writes the latest token into the JWT secret. The stack passes the JWT secret ARN to the contact flow.
Configure WebSocket Authentication for Agentforce Voice: When a call enters Amazon Connect, Amazon Connect opens a real-time connection to Salesforce and routes the call to the Agentforce agent. To secure the connection, Amazon Connect sends an API key (referred to by Salesforce as a JWT token) and stores it in AWS Secrets Manager. The token authenticates requests from your Amazon Connect instance. Authentication steps vary depending on whether you use Salesforce Voice with Telephony Providers.
Configure AWS Access: Confirm that your Amazon account can create resources for Agentforce Voice with Real-Time Streaming. Configure settings using your root user account or grant the required permissions to an IAM user or role.
Before You Begin: Complete these prerequisites before you configure Amazon Connect.
Configure Channel and Telephony Settings for Real-Time Streaming: Create a channel to manage inbound calls for the Agentforce agent. Configure your telephony system to create a VoiceCall record in Salesforce for every incoming call and transfer calls using real-time streaming. When necessary, transfer calls to a human service rep.
Register the A2A Application: Create an Agent-to-Agent (A2A) server application that points Amazon Connect at your public Agentforce WebSocket URL and JWT secret. Save the application ID and ARN, and confirm the application type, URL, authentication type, and JWT secret ARN.
Complete the Prerequisites: Before you connect an agent to your partner telephony system, complete these prerequisites.
Configure the Channel and Telephony Settings for SIP or Dynamic Routing: Create a channel that’s used to manage inbound calls routed to the Agentforce agent. If supported, configure your telephony system to create a VoiceCall record in Salesforce for every incoming call, and then transfer each call to the Agentforce agent using the SIP address or dynamic routing phone number. When needed, transfer the call to a human service rep.
Choose Your Telephony Provider for Voice-Enabled Agents: Route voice conversations to your voice-enabled Agentforce Service agents. Voice-enabled agents use Agentforce Voice features to provide the autonomous AI layer for your contact center. Agentforce Voice replaces static, hierarchical IVR logic with dynamic, intent-driven conversations.
Agent Actions: Actions are how agents get things done. Agents include a library of actions, which are the tools an agent can use to do its job. For example, if a user asks an agent for help with writing an email, the agent first selects a subagent, then it launches an action that drafts and revises the email and grounds it in relevant Salesforce data.
Add an Action to a Subagent in the Legacy Builder: To use an action with an agent, add it to a subagent. An action can be added to multiple subagents. When you add a subagent to an agent, Salesforce makes a copy of the subagent and its actions. Each agent and agent version has its own independent copy of subagents and actions.
Create a Custom Agent Action in the Legacy Builder: With custom actions, you can extend your agent with Salesforce functionality so that it’s ready to assist users with tasks specific to your business. Build custom actions using Apex, flows, prompt templates, and API integrations.
Best Practices for Agent Action Instructions: Action instructions tell an agent what your action does and when to use it. When you create a custom action, carefully plan and test your instructions to ensure that your action performs as expected.
Standard Agent Action Reference: Standard actions are included by default with your agents. Many standard agent actions are available to all users who have permission to access Agentforce. Other standard actions are built to work with specific clouds or products, so they require an additional license or permission. When you add a standard action to your agent, customize the action for your use case, including the underlying flow, prompt template, or other reference action.
Field Service | Assign Appointment to Service Resource for Field Service: Assigns the service appointment selected by the dispatcher to a service resource, in the gap identified in the service resource’s schedule on a specific date. This action is one of the actions provided by the Agentforce: Field Service Actions feature.
B2B Commerce | Assign Subscription: Assigns a selling model and proration policy to a product or bundle. Use this action to configure subscription billing behavior by specifying the exact selling model and proration policy names from the List Subscription Options action output. This action requires Commerce Subscriptions.
B2B Commerce | Attach Promotion to Campaign: Associates a promotion with a campaign for targeted marketing. Use this action to link promotions to a specific marketing campaigns, enabling you to organize promotional offers by campaign, track campaign effectiveness, and coordinate multi-channel marketing efforts with aligned promotional strategies.
B2B Commerce | Create Coupon: Generates a coupon code string for an existing promotion. Use this action to create redeemable coupon codes that customers can apply at checkout for promotions you've already configured, enabling you to distribute promotional offers through marketing campaigns, email communications, or customer service interactions.
B2B Commerce | Create Product: Creates a simple product with name, SKU, and optional description. Use this action to quickly add new products to your catalog with essential details, allowing you to expand product information later as needed. This action requires Commerce Subscriptions.
B2B Commerce | Create Product Bundle: Creates a static product bundle with optional child products and category assignment. Use this action to group related products together into a single purchasable bundle, streamlining the buying experience and enabling merchandisers to create curated product collections. This action requires Commerce Subscriptions.
B2B Commerce | Get Account Category Churn: Retrieves product categories that an account stopped purchasing from for a specific store. Use this action to detect category churn patterns by analyzing purchasing behavior over a 6-12 month period and comparing it to recent activity, enabling you to create targeted win-back campaigns that re-engage customers in abandoned product categories before they fully disengage. This action requires Salesforce Data Cloud.
B2B Commerce | Get Account Revenue Trend: Identifies accounts with declining year-over-year revenue for a specific store. Use this action to monitor account health and proactively reach out to at-risk customers by tracking revenue trends, order frequency, and average order value, helping you protect revenue and strengthen customer relationships before accounts churn. This action requires Salesforce Data Cloud.
B2B Commerce | Get At Risk Accounts: Identifies at-risk accounts by order recency and year-over-year revenue decline for a specific store. Use this action to prioritize customer retention efforts by surfacing accounts that show warning signs of disengagement through decreased ordering activity and declining revenue, enabling you to intervene early with personalized outreach, targeted promotions, or account check-ins to re-engage customers, protect revenue, and prevent full account churn before it's too late. This action requires Salesforce Data Cloud.
B2B Commerce | Get Bundle Status: Queries a bundle's full setup state and returns the checklist with suggested next steps. Use this action to guide bundle configuration by identifying incomplete setup tasks and recommended actions, ensuring all required components are properly configured before the bundle goes live. This action requires Commerce Subscriptions.
B2B Commerce | Get Customer Profile: Retrieves a 360-degree customer profile including revenue, ordering behavior, and top product categories for a specific account and store. Use this action to gain comprehensive insights into customer purchasing patterns and preferences, enabling you to personalize outreach, identify cross-sell opportunities, and make data-driven decisions about accuont management strategies. This action requires Salesforce Data Cloud.
B2B Commerce | Get Dead Stock Inventory: Identifies products that have not sold within a specified number of days for a given store. Use this action to proactively manage inventory by surfacing slow-moving or stale products, enabling you to implement targeted strategies such as promotional discounts, category reassignments, or inventory clearance to optimize stock levels and improve cash flow. This action requires Salesforce Data Cloud.
B2B Commerce | Get Order Pipeline: Provides an analysis of the returns fulfillment order pipeline, including the status breakdown, delayed orders, and on-time delivery rate for a specific store. Use this action to monitor returns processing efficiency and identify operational bottlenecks, enabling you to address delayed order proactively, improve customer satisfaction, and optimize your returns fulfillment workflow to meet service level expectations. This action requires Salesforce Data Cloud.
B2B Commerce | Get Product Status: Returns configuration state and checklist for a product. Use this action to guide product setup by identifying incomplete configuration tasks and recommended next steps, ensuring all required attributes and settings are properly configured before the product is published to your store. This action requires Commerce Subscriptions.
B2B Commerce | Get Product Trend: Returns month-by-month product revenue and units trend with overall growth classification for a specific store. Use this action to analyze product performance over time by tracking revenue and sales volume trends with optional product and category filters, enabling you to identify high-growth products, detect declining trends early, and make data-driven merchandising decisions about inventory, promotions, and product mix optimization. This action requires Salesforce Data Cloud.
B2B Commerce | Get Returns Analysis: Provides an analysis of product returns for a store, including return rates, top returned products, reason breakdown, and monthly trend. Use this action to identify quality issues, improve product descriptions, and reduce return rates by understanding which products are most frequently returned and why, enabling you to take corrective actions such as updating product information, addressing quality concerns with suppliers, or refining sizing guides to minimize future returns and improve customer satisfaction. This action requires Salesforce Data Cloud.
B2B Commerce | Get Revenue Summary: Provides store-level revenue metrics with period-over-period comparison for a specific store by querying Data Cloud. Use this action to monitor overall store performance and track growth trends by comparing current revenue against previous periods, enabling you to quickly assess store health, identify revenue acceleration or decline patterns, and make informed decisions about resource allocation, marketing investments, and strategic initiatives to drive business growth. This action requires Salesforce Data Cloud.
B2B Commerce | Get Segment Performance: Compares customer segments (by country or behavior classification) for a specific store, returning revenue, orders, customers, AOV, growth compared with prior period, and revenue share. Use this action to understand how different customer segments contribute to your business by analyzing performance across geographic regions or behavioral groups, enabling you to identify high-value segments, customize marketing strategies and promotions to specific customer types, and allocate resources effectively to maximize growth in your most profitable segments. This action requires Salesforce Data Cloud.
B2B Commerce | Get Top Performers: Identifies the top-performing products or customers by a selectable metric (revenue, units, frequency, growth) for a specific store. Use this action to understand what drives your store's success by ranking products or customers based on the metrics most relevant to your business goals, enabling you to focus inventory on best-sellers, reward and retain your most valuable customers, replicate successful patterns across your catalog, and allocate marketing resources to amplify your highest-performing areas. This action requires Salesforce Data Cloud.
B2B Commerce | List Subscription Options: Lists the available selling models and proration policies. Use this action to discover valid configuration options before assigning them to a subsciption product or bundle, ensuring you select the appropriate billing model and proration behavior that aligns with your business requirements and preventing configuration errors by referencing exact option names during product setup. This action requires Commerce Subscriptions.
B2B Commerce | Update Product Settings: Updates these fields for one or more product or bundles: category, entitlement policy, pricing, and inclusive pricing (applicable for bundles only). Use this action to efficiently manage product attributes at scale by applying updates across multiple product simultaneously using comma-separated product names, enabling you to quickly adjust pricing strategies, reorganize catalog categories, or modify entitlement policies without manually editing each product individually, saving time and ensuring consisitency across your product catalog. This action requires Commerce Subscriptions.
Marketing | Summarize a Campaign: Summarize a campaign with Agentforce in Marketing Cloud. Campaign details can include flow name, open rate, click through rate, sent, not sent, unsubscribed, opted out, and bounces per flow.
Security | Classify Security Risk: Identify and prioritize security risks and threats within Salesforce Security data. Provide clear and actionable remediation plans to guide admins on effectively responding to security issues.
Service | Close Case: This action lets customers retrieve case information and close cases.
Sales | Create Close Plan: Generates a sales plan to help sales reps close an opportunity by its target date.
B2B Commerce | Add Order to Cart: Creates a new order with the same items as a previous order. The new order is based on the previous order reference number. This action also provides a link for the user to complete the purchase.
Marketing | Draft a Campaign Preview: Creates a campaign preview that is generated as part of the brief. The action creates a series of campaign messages, or steps, such as email or SMS. Use in conjunction with the Save Campaign action.
Sales | Create a Label: Creates a personalized label to contextually group and organize records.
Sales | Create a To-Do: Create tasks for follow-up or assignments to complete, helping sellers manage their work efficiently.
Field Service | Create Appointment List Filter for Field Service: Creates a filter for the service appointment list in the dispatcher console. If a user triggers QueryRecords to retrieve service appointments with certain criteria and then asks to see the service appointments in the appointment list, the action creates a filter in the Appointment List containing the service appointments returned by Agentforce. This action is one of the actions provided by the Agentforce: Field Service Actions feature
Service | Create Case with Enhanced Data: Creates a detailed support case with a structured summary, relevant conversation history, and a linked chat transcript when triggered from supported messaging channels. This context helps agents resolve issues efficiently and avoid repeated interactions.
Financial Services | Create Interaction: This action lets relationship managers create interaction records for meetings and conversations with clients or partners. The action identifies the account name from the conversation. If the account name is unknown, the action prompts the user to provide either the account ID or interaction ID before creating an interaction record.
Financial Services | Create Interaction Summary: Log meeting notes, next steps, and related records based on meeting discussion with interaction summary records. The action identifies the account name from the conversation. If the account name is unknown, the action prompts the user to provide either the account ID or interaction ID before creating an interaction summary record.
Financial Services | Create Opportunity Based on Meeting Notes: Creates an opportunity record based on meeting notes captured during interaction with the client. The action identifies the account name from the conversation. If the account name is unknown, the action prompts the user to provide account ID to log an opportunity.
Marketing | Draft a Campaign Brief: Creates a draft of the campaign brief based on user-provided campaign details in the Agentforce panel. Use in conjunction with the Save Campaign Brief action.
Agentforce Platform | Draft or Revise Email: Creates or revises a draft of an email to a recipient based on the user’s request and available email types. Email drafts include a recipient, subject, and email body. Email drafts can be sent through the Salesforce email composer or copied to another email client.
Service | Enhance Product Description: Analyzes product descriptions against customer reviews to identify discrepancies, align features, and suggest improved descriptions.
Sales | Explore Conversation: Answers a user’s questions about a voice or video call based on the contents of the call transcript. For example, if the user enters, “What was the customer sentiment on the call?” in the Einstein panel, the action delivers an answer based on the customer’s language in the call transcript.
Agentforce Platform | Extract Fields and Values from User Input: Extracts fields and their corresponding values from a record based on the user's request. The output of this action is used as the input for the Update Record action. For example, when a user wants to edit a field on a record, the Extract Fields and Values from User Input action is triggered first to retrieve the field and its value, and then the Update Record action saves the updated extracted value to the record.
Sales | Find Past Collaborators: Leverages insights from Einstein Activity Capture to identify individuals with past collaborative ties, aiding in securing introductions to relevant parties in ongoing or future deals.
Sales | Find Similar Interactions: Searches interaction records by channel or related terms and generates an engagement summary for specified or all channels, including case, chat, email, surveys, and voice.
Sales | Find Similar Opportunities: Searches for and returns a list of opportunities that are similar to, but not duplicates of, a specified Salesforce opportunity. An explanation of the similarities is returned with each opportunity record.
Financial Services | Get Client Life Events: The Get Client Life Events action provide wealth advisors with a list of past and upcoming life events within Salesforce Financial Services Cloud.
Financial Services | Review Asset Allocation: The Review Asset Allocation action provides wealth advisors with a comparison of an account’s current portfolio asset allocation with their target allocation.
Financial Services | Summarize Account Financial Details: The Summarize Account Financial Details action provides wealth advisors with real-time summary of an account’s financial profile within Salesforce Financial Services Cloud.
Financial Services | Summarize Household Financial Details: The Summarize Household Financial Details action provides wealth advisors with a real-time summary of a household’s financial profile within Salesforce Financial Services Cloud.
Financial Services | Summarize Financial Plans and Goals: The Summarize Financial Plans and Goals action finds and retrieves financial plans and financial goals records based on user input and specific conditions such as field values. This action identifies the correct records and object.
Financial Services | Summarize Portfolio Performance: The Summarize Portfolio Performance action provides wealth advisors with a summary of an account’s portfolio performance within Salesforce Financial Services Cloud.
Sales | Get Activities Timeline: Retrieves a list of all CRM activities associated with a record during the specified time frame. The list can include past and future activities. The action is exposed through Flows, making it available for automation.
Sales | Get Activity Details: Provides a brief summary of an activity, including an overview of its content (such as an email or a call summary) and other relevant details. Valid activity types are calls, emails, events, or tasks.
Field Service | Get Appointments to Fill Gaps for Field Service: Searches for gaps for a specific service resource on a specific date and returns a list of suitable appointments in ranked order, considering business rules and objectives. This action is one of the actions provided by the Agentforce: Field Service Actions feature.
Sales | Get Conversation Intelligence: Gets intelligence information about a voice or video call, including any insights and a conversation summary if one is available.
Sales | Get Conversation Transcript: Gets a transcript for a voice or video call. This can be used to automate updates to records based on conversation data.
Marketing | Get Distributed Marketing Email Templates: Retrieves relevant Distributed Marketing and Alerts email templates so non-marketing users can personalize and send approved campaigns. Each template includes the campaign name, subject line, preheader, content, and campaign start and end dates.
Sales | Get Forecast Guidance: Gathers information about how a sales rep is tracking toward meeting their Committed and Most Likely quota for the current forecast period. The output includes the seller’s forecast amount, a list of any deals at risk, and why they’re considered at risk.
Financial Services | Get Interactions By Date Range: Gets the ten most recent interaction records that are associated with the client account. Filters these interaction records to return only those that occur within a given date range.
Marketing | Get Marketing Engagement Activities: Retrieves engagement history for up to five leads or contacts to help non-marketing users personalize outreach and follow-ups. Each result includes recent activities such as email opens, clicks, and web visits with timestamps.
Service | Get Order by Order Number: The Get Order by Order Number action returns the order information associated with a given contact ID and order number.
Security | Get Policies by Object: Use the Get Policies by Object action to retrieve a list of all policies in your Policy Center setup that contain a specific object.
Security | Get Policies by Policy Type: Use the Get Policies by Policy Type action to retrieve a complete list of every policy of a specific policy type in your Policy Center setup.
Security | Get Policy Details: Use the Get Policy Details action to retrieve information about a policy in your Policy Center setup, such as the objects included in the policy.
Sales | Get Product Pricing: Gets the pricing information of a product, including relevant historical sale price data from previous won deals involving the same product.
Marketing | Get Recommended Email Templates: Recommends approved email templates—ranked by relevance—for a lead, contact, or person account based on their engagement history. Each recommendation explains why it was suggested, helping you choose the right template.
Agentforce Platform | Get Record Details: Retrieves the details of a record, including object fields and values, records from related lists, and associated tasks and events.
Security | Get Security Alerts: Cut through Salesforce notification overload by providing conversational, proactive security posture alerts with the Get Security Alerts agent action.
Security | Get Security Metric Data: Translate your security posture into natural language with the Get Security Metric Data agent action. Ask questions, get answers, and spot threats before they strike – all through simple conversation.
Agentforce Platform | Get User Memories: Retrieves the collection of memories stored for a specific user and AI agent, for use in personalizing agent responses.
Health | Generate EMR Data Fetch Payload: Generates the input payload required to get Fast Healthcare Interoperability Resources (FHIR) data from Electronic Medical Record (EMR) systems.
Health | Generate Patch FHIR Payload: Generates the Patch FHIR input payload required to update resources, such as Patient and Medication Request.
Sales | Identify Contact Role: Identifies the role, significance, or impact of a contact in influencing the opportunity or account status. This action also assesses how that contact affects the outcome of a deal, initiative, or account.
Sales | Identify Key Contacts: Identifies key contacts associated with an account or opportunity. A key contact is someone who can influence, obstruct, impede, or decide a deal, or is involved in the outcome of the deal.
Agentforce Platform | Identify Object by Name: Interprets the user’s input to determine which object the user is referring to and then returns a list of matching object names so other actions can be taken. For example, if the user enters, “List the opportunities for the Acme account” in the Agentforce panel, the action identifies that the user is requesting information related to the Account object and Opportunity object.
Agentforce Platform | Identify Record by Name: Searches for Salesforce records by name and returns a list of matching record IDs. For example, if a user enters, “Show me the Acme records” in a Salesforce org with multiple records related to “Acme,” the action can return the Acme account, the Acme - 20k opportunity, and the Acme - 50k opportunity.
Marketing | Identify Top Engaged Unified Individuals: Retrieves a summary of the most engaged individuals to help non-marketing users with strategic follow-ups. With scoring configured, the summary also shows the individual's scores.
Sales | Label a Record: Assigns personalized labels to records. Users can tag records based on meaningful attributes, retrieve those records as a group, and act on them.
Sales | Log a Call: Logs a call with a customer as a task, which can include additional details about the call, such as the date, time, and purpose. The logged call appears in the seller’s Activity Timeline.
Agentforce Platform | Manage User Memory: Uses generative AI to create, update, or delete memories about a specific user's information and preferences, based on a conversation with an AI agent.
Marketing | Create Section with Content: Creates a section in an email or landing page for campaigns and initiatives. Users can specify the number of columns in the section and the components to be included in each column, such as Heading, Paragraph, or Button. Users can also ask the agent to create content for the components in the section.
Marketing | Draft Content: Creates or revises copy for any text-based property for campaigns and initiatives, including paragraph, heading, email subject line and preheader, and SMS message. Users can revise generated text by entering a different key message or target audience or by selecting a different tone.
Marketing | Get B2B Marketing List Memberships: Retrieve a list of marketing lists a prospect is a member of, whether they're subscribed to the lists, and whether the lists are used in nurture programs.
Marketing | Get B2B Marketing Lists: Retrieve all marketing program lists for a prospect. To retrieve all email preference center subscription lists, use the isPublic filter. To retrieve all nurture programs, use the isSalesforceVisible filter.
Marketing | Upsert B2B Prospect: Create or update a prospect record, including its profile, score, and owner. To use this action with an agent, include an Email, Prospect ID, Lead ID, or Contact ID.
Marketing | View B2B Prospect: Retrieve a prospect record, including its profile, score, and owner. To use this action with an agent, include an Email, Prospect ID, Lead ID, or Contact ID.
Personalization | Get Recommendations: This agent action generates product and content recommendations tailored to the user’s preferences, behavior, and interactions. The recommendations are based on the current context or specific user query.
Personalization | Understand User Intent: Understands a user's natural language query to extract their intent, including the item and relevant filter criteria, which results in more personalized recommendations.
Personalization | Get Context: Provides an agent with personalized context from Data Cloud to help it select the most appropriate next action in customer interactions.
Sales | Prioritize Opportunities: Identifies a rep’s top sales opportunities and ranks them in order of priority based on specific instructions provided by the user or available opportunity data. An explanation for each opportunity’s rank is also returned.
Agentforce Platform | Query Records: Finds and retrieves Salesforce records based on the user’s request and specific conditions, such as the values of fields.
Sales | Research Account: Conducts external research and generates a summary of findings about the company on a business account. The output depends on the user’s research question and can include information about the company’s structure, business model, revenue, and competitors.
Service | Reset Password: This action sends a password reset link to the email address associated with the given contact record.
Service | Reset Secure Password: This action sends a password reset link to the email address associated with the given verified contact ID.
Financial Services | Restructure Meeting Notes: Organizes raw meeting notes into key points, decisions, and action items for clear and easy tracking. Gathers recently modified opportunity records and account plan objective records associated with the account plan record for an account and summarizes the meeting highlights.
Sales | Review My Day: Provides an overview of the seller’s daily agenda, including pending tasks, meetings, leads, opportunities, and how the seller is tracking against their sales quota. Only the top three leads and opportunities that require your attention are shown. The leads are sorted in descending order of the annual revenue. The opportunities are sorted in ascending order of the Close Date. A maximum of three open tasks that the seller owns are shown in descending order of priority.
Sales | Suggest Next Step for Opportunity: Identifies and suggests a value for the Next Step field on an opportunity. This action is intended for use with Agentforce Pipeline Management.
Sales | Suggest Stage for Opportunity: Identifies and suggests the appropriate stage for an opportunity. This action is intended for use with Agentforce Pipeline Management.
Marketing | Save Campaign: Saves the campaign and its draft components. Use in conjunction with the Create a Campaign from a Brief action.
Marketing | Save Campaign Brief: Saves the brief information, including name, description, key message, and target audience. Use in conjunction with the Draft a Campaign Brief action.
Scheduler | Collect Appointment Details for Scheduler: Collects the details required to schedule a service appointment. For example, if the user wants to schedule an appointment, this action asks for details such as the date, subagent, service resource, and so on.
Scheduler | Get Appointment Time Slots for Scheduler: Get the available slots for the appointment. This action is invoked when an appointment is booked or modified. Call this action with a related action that can provide the input it requires.
Agentforce Platform | Search the Web: Retrieves real-time or external information that isn’t available in Salesforce CRM. Use it for questions about news, market trends, competitor details, and product comparisons, or when Salesforce answers need added context from the web.
Agentforce Platform | Send Email with Verification Code: Send a verification code to the customer’s email to verify their identity. The action requires the customer’s username or email as input. If the provided email is valid, a verification code is sent to the customer.
Sales | Send Meeting Request: Creates an email draft to request a meeting with the recipient, with three possible time slots, a link to view the slots on a new page, or a link to the sender’s calendar where they can pick a slot to schedule a meeting.
Slack | Send a Slack Message to a Channel: Sends a message to a Slack channel you specify. Messages in public channels are visible to anyone in Slack. Messages in private channels are only visible to members of the channel.
Agentforce Platform | Speech to Text: Converts an audio file into text and returns the transcript for use in agent conversations or downstream actions.
Life Sciences | Summarize Patient Response for Benefits Reverification: Analyzes the patient's response for benefit reverification email and summarizes the changes to the benefits verification request, including personal, drug, healthcare provider, pharmacy, and insurance details.
Service | Summarize Product Reviews: Summarizes customer reviews and feedback to identify the top products from the specified products and show positive, negative, mixed, or neutral sentiment.
Agentforce Platform | Summarize Record: Generates a summary of a record. The output depends on the record type and the prompt templates the user has access to.
Field Service | Summarize Scheduling Issues for Field Service: Creates a summary of the total number of appointments that require immediate attention. The summary is divided according to exception categories. The agent action creates a filter for each category in the appointment list. This action is one of the actions provided by the Agentforce: Field Service Actions feature. The categories used by the reports are customizable.
Security | Summarize User Activity: Identify potential security threats and anomalies by analyzing user activity data. Use the generated risk analysis summary and proposed remediation steps to effectively respond to security issues.
Service | Update Customer Contact: This action updates fields on a customer’s contact record, such as the email address, phone number, or postal address.
Agentforce Platform | Update Record: Updates one or more fields on a record. To retrieve the fields and values the user wants to change, the Extract Fields and Values from User Input action must be triggered before the Update Record action.
Service | Update Verified Contact: This action updates fields on a verified customer’s contact ID, such as the email address, phone number, or postal address.
Agentforce Platform | Verify Customer: Verifies the customer’s identity based on the one-time passcode (OTP) generated and emailed by the Send Email with Verification Code action. When a user initiates verification, the system indicates that a verification code will be sent if the provided email or username is valid, and prompts the user to enter the code. If the user verification is successful, it outputs the verified user or contact ID, which can be used in filters and as input to other actions. If the verification is unsuccessful, it prompts the user to retry.
Service | Update Omni-Channel User Configuration: Supervisors can change reps' presence statuses, queue assignments, skill assignments, and skill levels for Omni-Channel users. For example, to decrease wait time, increase customer satisfaction, and increase operational efficiency, the supervisor can temporarily assign idle agents to the busiest queues. Using the conversational interface, they explicitly identify skill levels by number and presence statuses, agents, skills, and queues by name. A supervisor can modify only the Omni-Channel users, queues, and skills that are visible based on their supervisor configuration.
Add the Search the Web Action to a Prompt Template: Add the Search the Web data action to a prompt template so your prompt can include current information from the web, such as current events, industry trends, and product information.
Legacy Agentforce Analytics: Legacy Agentforce Analytics uses Data 360 to help you gain insights into how your agents are performing. The insights in your customizable dashboard and reports can help you determine whether you want to change anything to drive adoption or adjust subagents and actions to improve user satisfaction.
The Building Blocks of Agents: Get to know the key components of agents, including subagents and actions, data, connections and channels, and the reasoning engine.
Customize Excluded Domains for the Search the Web Agent Action: Limit web search results by excluding specific websites from the Search the Web agent action. Exclude the domains you don't want so your agents skip those sources and retrieve more relevant information.
Design and Implement Agents: Agentforce is the agent-driven layer of the Salesforce Platform, and it helps you deploy AI agents that work side-by-side with your employees to extend your workforce and serve customers 24/7. Agentforce includes a set of tools to create and customize agents for use cases across sales, service, marketing, commerce, Slack, and more.
What Are Agents?: Agents are goal-oriented, autonomous AI assistants that perform tasks and business and make data-driven decisions. Learn more about agents and how they compare to Einstein Bots.
Agentforce Standard Asset Reference: Get to know Salesforce standard agent actions and more with our comprehensive reference libraries, so you can quickly add powerful out-of-the-box functionality to your AI agents.
Agentforce Analytics Reports: Agentforce Analytics is based on Data 360 reports that can help you gain insight into the performance of your agents. The reports show you who’s using your agents, how successful they are, and what your adoption level is, among other things.
Activate or Deactivate Your Agent: Activate your agent to make it available to your customers or employees. When an agent is deployed to one or more channels, including the Agentforce panel, activating the agent makes it immediately available to your users.
Enable Enhanced Event Logs: Event logs capture the events and user messages in an agent session to help you test and troubleshoot your agent. In supported agent types, enable enhanced event logs to store conversation data and view all session activity in one place. When enhanced event logs aren't enabled, events associated with agent and user messages (such as Message sent: and User response:) are logged, but the messages themselves aren't included in the event.
Define Agent-Level Instructions and System Messages: Agent-level instructions provide your agent with guidance on how to interact in all situations, including any rules it should follow. Your agent sends system messages, such as welcome and error messages, when your users begins chatting with an agent and if they encounter system errors.
Update Language Settings: Change the language and tone of your agent’s messages to better suit your company’s brand.
Subagents in the Legacy Builder: A subagent is a particular job an agent can do and an essential element of an agent’s reasoning. A subagent contains actions, which are the tools available for the job, and instructions, which tell the agent how to make decisions. Collectively, the subagents assigned to your agent define the capabilities your agent can handle. Salesforce provides a library of standard subagents for common use cases, and you can create custom subagents to meet your users’ specific business needs.
Add a Subagent to Your Agent in the Legacy Builder: You can create a custom subagent from scratch or add a standard subagent from the asset library. Consider best practices for creating a subagent as you decide. When you add a subagent to an agent, Salesforce makes a copy of the subagent so that each agent and agent version has its own independent copy.
Add a Subagent from the Asset Library in the Legacy Builder: With standard subagents, your agents can start handling common business use cases quickly. You can create subagents and actions that are tailored to your business needs. When you add these subagents and actions to an agent, the agent gets its own, independent copy. Changes to template subagents and actions within the Asset Library aren't automatically sync'd to subagents and actions within your agent. Changes made to one agent's subagents or actions don't affect another agent's subagents or actions.
Edit a Standard Subagent in the Legacy Builder: Edit an out-of-the-box subagent to make sure it meets the needs of your specific use case and respects the right guardrails for your business.
Best Practices for Subagent Instructions in the Legacy Builder: Subagent instructions set guidelines for agent behavior and give an agent the context it needs to do its job. Review best practices for writing subagent instructions to reduce latency and help your agent respond accurately and consistently.
Standard Agent Topic Reference: Standard topics are topics available out of the box in the asset library. Many standard topics are available to all users who have permission to access Einstein Copilot. Some standard topics are built to work with specific clouds or licenses, so they require another license or permission.
Agent Topic: Commerce Promotions: Use the Commerce Promotions topic to create promotions or get promotion suggestions and templates for a Commerce store.
Agent Topic: Field Service Dispatcher Actions: Use the Field Service Dispatcher Actions topic to perform tasks in the dispatcher console. Display a summary of today’s scheduling issues and show them in the appointment list for subsequent follow-up actions. Search for appointments and create filters in the appointment list. Find gaps in a service resource’s schedule, find the best appointments to fill those gaps, and schedule the appointment that is selected.
Agent Topic: Insights Business Objectives: Use the Insights Business Objectives topic to get recommendations to improve Key Performance Indicators (KPIs) based on your store's data.
Trust and Agentforce: With Salesforce, it’s easy to build safe AI agents because agents are designed to help mitigate the risks associated with generative AI technology.
Utterance Analysis: Utterance Analysis lets you quickly see how Agentforce (Default) users are using agents, what they’re requesting, and whether the agent was able to handle these requests. These customizable reports can help you refine your subagents and actions so that your agents respond more effectively and accurately.
Configure Search Settings in Salesforce Classic: Enable document content search, CJKT search optimization, sidebar search auto-complete, and more. Configure the lookup settings and the number of search results per object and lookup settings.
Customize Conversation Recommendations for Agentforce Employee Agents in Agentforce Builder: Customize conversation recommendations to suggest prompts your employees can use with your Agentforce Employee Agent, so they can select a prompt to start a conversation instead of typing from scratch. Conversation recommendations include welcome recommendations that appear on the welcome screen when the Agentforce panel opens, and recommendations that appear after agent responses during a conversation.
Customize Search Layouts: Search layouts let you determine what users see when Einstein Search returns results. For each unique user profile, you can create search layouts for standard and custom objects, ensuring the layout shows users what’s most relevant to them. Only objects with customizable layouts support profile-specific layouts. The search layout that you configure applies to global and lookup searches.
Choose A Data Source for Your Library: When you configure a data library, you can add knowledge articles, uploaded files, or custom retrievers as a data source. The data source is indexed on Data Cloud where a search index and retriever are automatically created for you. When integrated with AI agents, the data library’s search index and retriever are used by the Answer Questions with Knowledge action.
What Are Data Libraries?: Agentforce Data Libraries enhance the accuracy of AI features like Agentforce Agents by connecting them to your trusted data sources. Data libraries use unstructured or semi-structured data so that they can take large sets of information that exist on the web, in documents, or as large text inputs in fields, and turn that into more useful, searchable information.
Use a Custom Retriever: Use custom retrievers to connect to more types of data sources in Data 360 including ensemble retrievers, or, connect companion orgs to data libraries and search indexes configured in your home org.
Use Uploaded Files: Select the files to upload for your data library to index.
Learn About Data Libraries: Understand how data libraries work and learn best practices for managing your content to use with generative AI.
Agentforce Data Library: Improve accuracy, add personalization, and build trust in gen AI responses when you assign data libraries to your Agentforce features. Agentforce Data Library uses grounding to index your knowledge articles and fields, your file uploads, or web sources, so that AI agents know which information to base responses on. With this index, AI agents can ensure the accuracy of LLM responses against your organization’s information so that you get the best results.
Prepare Your Source Content: Get the most out of your source content by preparing it so that it’s ready for AI and human consumption. Discover best practices for structuring your content, tailoring to your audience, adding important details, and maintaining accuracy over time.
Test Data Libraries in a Sandbox: A sandbox is a copy of your Salesforce org that you can use for development, testing, and training without compromising the data and applications in your production org. Agentforce Data Library is supported in all the sandbox types: Developer, Developer Pro, Partial Copy, and Full Copy.
Use Web Search: Agentforce Data Library doesn't support new web-search data libraries. You'll still be able to view and use your existing web search data libraries.
Sales Cloud Einstein: Supercharge your team’s sales productivity at every step of the sales process with key predictions, intelligent recommendations, and timely automation. Sales Cloud Einstein is your own data science department that learns from your team’s sales activities and CRM data. Einstein helps you identify the best leads, convert opportunities more efficiently, and retain customers with ease. Sales Cloud Einstein also includes the Sales Analytics app and Inbox.
Einstein Automated Contacts: Spend even less time on data entry. Einstein Automated Contacts uses email and event activity to find new contacts and opportunity contact roles to add to Salesforce. Depending on how your Salesforce admin set up the feature, Einstein suggests the new data or adds it for you.
Review the CRM Analytics Automated Contacts Dashboard: The CRM Analytics Automated Contacts dashboard provides Salesforce admins and sales managers with details about how effective Einstein is at finding and adding new contact records to Salesforce.
Bring Intelligence to Your Daily Workflow with Sales Cloud Einstein: Take Einstein to the next level with intelligence that impacts daily decisions. Einstein Insights gives sales reps relevant updates and follow-up reminders to help them win more deals and nurture relationships with accounts. Einstein Scoring helps reps prioritize leads and opportunities so they can focus on what’s most likely to convert and close.
Review the CRM Analytics Lead Scoring Dashboard: The included CRM Analytics Lead Scoring dashboard provides insights into which lead characteristics result in the most conversions. See how well lead scores correlate to conversion rates, monitor the average lead score of each lead source, and more.
Einstein Generative AI for Sales: Einstein Generative AI for Sales connects your sales data with the generative capabilities of artificial intelligence (AI) and large language models (LLMs). With Sales Emails, your sales team can generate personalized emails to send to contacts and leads. Your team gets precious time back in their day by getting nearly-there drafts, already contextualized and grounded in your Salesforce data. And because trust is our #1 value, you can rest easy knowing that your data is secure.
Considerations for Einstein Generative AI Sales Emails: Sales Emails lets your team use Einstein generative AI to create personalized and contextualized emails to send to your contacts and leads. But whether you’re an admin setting things or a sales rep powering through your day, keep these considerations in mind.
Understand How Einstein Generative AI Creates Sales Emails: Sales Emails lets busy sales reps to speed up and expand deals through personalized and informed content that buyers love. Einstein generative AI generates personalized emails for contacts and leads, and uses Einstein to personalize and contextualize the emails, a method known as grounding. Learn about email types, how your data is used, and how Salesforce tracks usage.
Set Up Sales Emails: Get ready to add Einstein Sales Emails to your sales team’s toolbox. Turn on Sales Emails in Setup and then assign a permission set to your sales reps. If you use Gmail or Outlook email integrations, add the Einstein Sales Emails component to customized email application panes. Then your sales reps can use Einstein where they normally work.
Build a Smarter Pipeline with Einstein Forecasting: Take the guesswork out of forecasting. Einstein Forecasting uses AI technology to bring more certainty and visibility to your forecasts. Improve forecasting accuracy, get forecast predictions, and track how sales teams are doing.
Enable Einstein Forecasting: Provide your forecast managers with AI-powered intelligence that improves forecasting accuracy, predicts results, and tracks how sales teams are doing.
Understand How Sales Cloud Einstein Works: Sales Cloud Einstein might seem like magic, but it’s not. It’s built with science. Data science. Learn more about how Sales Cloud Einstein features use your data to bring intelligence to Salesforce.
Einstein Lead Scoring: Use AI to score your leads by how well they fit your company’s successful conversion patterns. Let your sales team prioritize their leads by lead score. See which fields influence each lead score most.
Einstein Opportunity Scoring: Let artificial intelligence help you and your team focus on the right opportunities so you can close more deals. Each opportunity is given a score, from 1 to 99, which is available on opportunity records and list views. If you use forecasts in Salesforce, opportunity scores are also available on the forecasts page. Plus, use scores with reports, Process Builder, and workflows. Einstein Opportunity Scoring is available to users with or without a Sales Cloud Einstein license.
Create Custom Report Types for Einstein Opportunity Scoring: Run reports that show how Einstein Opportunity Scoring impacts your business or what factors have the biggest affect on your deals. First, set up custom report types so that sales managers and other teammates can create the reports they need.
Increase Productivity with Sales Cloud Einstein: Free up sales reps’ valuable time and increase productivity. When reps connect their email and calendar to Salesforce, Einstein Activity Capture adds email and events to related Salesforce records. This busy work is done automatically, so reps don’t have to.
Sales Cloud Einstein Scoring: Always know what to work on first. Einstein Scoring helps sales reps prioritize leads and opportunities so they can focus on what’s most likely to convert and close.
Set Up Sales Cloud Einstein: Use the Sales Cloud Einstein Setup Assistant to get the targeted guidance you need for setting up the Sales Cloud Einstein features you want.
Disable Einstein Lead Scoring: If your company decides to turn off Einstein Lead Scoring, take steps to avoid errors with CRM Analytics Data Sync.
Set Up Einstein Activity Capture for Sales Cloud Einstein: Set up Einstein Activity Capture so that emails and events are automatically associated with related Salesforce records. Sales reps must connect their email and calendar to Salesforce.
Enable Einstein Automated Contacts: Help reps spend even less time on data entry. Einstein Automated Contacts uses email and event activity to find new contacts and opportunity contact roles to add to Salesforce. Choose whether Einstein suggests the new data, which reps can add with just a couple of clicks, or adds it automatically.
Troubleshoot Sales Cloud Einstein Setup Errors: When setting up Sales Cloud Einstein features, several important steps occur behind the scenes. If one of the steps isn’t successful, one or more features can’t be enabled. There are several ways to troubleshoot setup issues.
Enable Einstein Lead Scoring: Give your sales team access to scores that help them prioritize leads. Turn on Einstein Lead Scoring, and then select a lead conversion milestone to use, which leads to score, and which lead fields to consider during scoring.
Enable Einstein Opportunity Scoring: Give your sales team access to scores that help them focus on the right deals. When you set up Einstein Opportunity Scoring, you choose whether to have Einstein consider all opportunity records and opportunity fields or only a subset. If Einstein Opportunity Scoring is on by default, make sure that scores appear where you want, such as your customized opportunity page layouts and public list views.
Prepare for Sales Cloud Einstein: Before you set up Sales Cloud Einstein feature, review information about data requirements and other considerations. Learn about sandbox support and how to run the readiness assessor.
Select Who Can Use Sales Cloud Einstein: The Sales Cloud Einstein Included and Sales Cloud Included Bundle standard permission sets include the permissions for Sales Cloud Einstein features. The permission sets also include some CRM Analytics features, such as dashboards and the ability to analyze your report data using Einstein Discovery for Reports. Assign a permission set to users.
About Einstein Generative AI: Learn more about Einstein generative AI and trust, terminology, and considerations for use.
Agent Analytics: Analyze your AI agents’ performance within user and agent sessions. Agent Analytics is built on the unified Session Tracing Data Model (STDM), which captures every event and interaction within a session and logs it as individual events. You can query the logged events to gain insights and optimize the Agent.
About Agent Analytics: Analyze the performance of your AI agents. Agent Analytics is a tool specifically designed for comprehensive session analysis. Built on the unified Session Tracing Data Model (STDM), it captures every turn and event occurring in and across all your agent sessions. You can then query the logged events to gain insights and optimize agent performance.
Agentforce Observability Dimension Formulas: Use these formulas to understand how Agentforce Observability calculates dimensions in the Agentforce Analytics Foundations semantic data model.
Agentforce Observability Measure Formulas: Use these formulas to understand how Agentforce Observability calculates measures in the Agentforce Analytics Foundations semantic data model.
Set Up Agent Analytics: Set up Agent Analytics to gain insights into agent performance, track conversation metrics, and optimize your agents with comprehensive dashboards and session insights.
Troubleshoot Agent Analytics: If you’re unable to see the information, models, dashboards, or metrics that you expect, try these solutions.
Get Insights with Agent Analytics: Monitor agent performance in Agentforce Studio with Agent Analytics dashboards, then open Sessions & Intents to audit transcripts, compare intents, and address low-quality sessions.
Customize Agentforce Observability: Customize Agent Analytics when the standard metrics don't fully represent your business processes. Use scorer results and the analytics semantic data model to refine metrics and surface them in dashboards.
Agentforce Observability Data Model: The Agentforce Observability data model combines Agent Analytics and Agent Optimization data with Agentforce Session Tracing. It includes session trace logs, intents, LLM calls, feedback, and Trust Layer metrics for analyzing agent performance.
Einstein Generative AI Analytics: Collect and store Einstein generative AI data in Data 360. Enhance your generative AI solutions with insights gained from dashboards and reports.
Monitor and Analyze Einstein Service Replies: Evaluate and track the effectiveness of Einstein Service Replies in your organization by creating custom report types that connect case and messaging session data with generative AI audit and feedback data.
Verify Masked Data: After you have configured data masking to in the Einstein Trust Layer Setup, you'll want to verify that sensitive data is masked and also monitor it over time.
Review Toxicity Scores: Toxicity scores measure the level of potentially harmful language detected by the Einstein Trust Layer in large language model (LLM) responses. To monitor toxicity trends in your Salesforce org, use Einstein Generative AI Audit and Feedback Data.
Audit Trail: Track the use of generative AI in your Salesforce org and ensure that AI usage complies with your security, privacy, regulatory, and AI governance policies.
Batch Models: Use Prompt Template Batch Processing to generate large quantities of responses for prompt templates asynchronously.
Billing Considerations for Data Prism: Data Prism uses Data 360 to query sample data and Einstein Requests to call generative AI. Data Prism is called by the Agentforce QueryRecords action and the Data 360 Einstein Segment creation feature. Use of Data Prism impacts the consumption of credits used for billing in these usage types.
Manage Salesforce Access to Customer Data: Choose whether to allow Salesforce to access Customer Data in your org. If shared, Salesforce has your consent to use Customer Data for specific purposes.
Einstein Audit and Feedback Data: Collect and store Einstein generative AI audit and feedback data in Data 360 and enhance your generative AI applications with feedback analytics, prompt adjustments, and LLM tuning.
About Generative AI Audit and Feedback Data: Use generative AI audit data (also known as audit trail) to track Einstein Trust Layer functionality such as data masking and toxicity detection that help ensure the safety and accuracy of the generated response.
Implement Data Governance Permissions for Audit and Feedback Objects: Audit and feedback data can contain sensitive information, such as personally identifying information (PII), that you want to redact or hide from unauthorized users. Use Data Cloud Data Governance to protect sensitive data by setting up dynamic data masking and field-specific data exclusion policies. Grant access to the impacted Data 360 objects according to an access policy.
Data Model for Generative AI Audit and Feedback: Understand and analyze your Einstein generative AI audit and feedback data by using the appropriate data model objects (DMOs). Einstein generative AI Audit and Feedback data is streamed to a data lake object (DLO) in Data 360 and mapped to data model objects (DMOs).
Set Up Einstein Generative AI Audit and Feedback: To collect Einstein generative AI audit and feedback data, turn it on in Einstein Setup. For orgs with only one dataspace, it’s turned on by default when Agentforce is turned on.
Access Einstein Generative AI Audit and Feedback Data Reports and Dashboards: Access the Einstein generative AI audit and feedback reports and dashboards directly from Data 360, after turning on the Einstein generative AI data collection and storage. You can access the Einstein generative AI and feedback data dashboard.
Set Up Email Notifications for Specific Types of Audit and Feedback Data: Use pre-built flows to automate the process of sending email notifications for specific content quality issues or user feedback thresholds or conditions. For example, you can be alerted if the percentage of toxicity in the responses is over the specified threshold.
Large Language Model Support: Understand supported large language models (LLMs) from multiple providers for embedded features, such as Prompt Builder. Identify the Salesforce-managed models that are available out of the box. Learn how you can bring your own model (BYOLLM) by using AI Models (formerly Einstein Studio).
Large Language Model Limits: Understand limits for supported large language models (LLMs) from multiple providers for embedded features, such as Prompt Builder. Limits for each model include requests per minute and token limits.
Large Language Model Multimodal Support: Salesforce-managed models have different levels of support and limits for including JPG, PNG, or PDF files in a model request.
Salesforce-Owned Models: Salesforce AI Research creates, trains, and fine tunes models to address specific Salesforce use cases. These models are hosted on AWS within the Salesforce Trust Boundary.
Large Language Model Data Masking: The Einstein Trust Layer identifies and masks sensitive information in the prompts before sending them to the large language model (LLM). The Einstein Trust Layer uses data masking to protect sensitive data such as personally identifiable information (PII) from being exposed to external LLMs.
Select What Data To Mask: Select which data you want to mask and prevent exposure of sensitive data to the large language model (LLM). These settings are applied to your Salesforce org.
Prepare for Model Deprecation and Rerouting: This document provides guidance on retesting Salesforce AI implementations during model deprecation and rerouting. Deprecation and rerouting are considered temporary. If your model is in one of these states, consider switching to a new model.
Manage Model Provider Access: Choose which Large Language Model (LLM) providers to allow or not allow in your organization. When access to a model provider is turned on, you can use its language learning models (LLMs) in agents, prompt templates, APIs, and other features in generative AI solutions. Turn off a model provider to block access to its models in your org.
Knowledge/RAG Quality Data and Metrics: Monitor knowledge retrieval quality scores for your Agentforce and generative AI solutions in production and sandbox environments. Use built-in dashboards and reports to track run-time performance, view trends, and identify areas to improve. Retrieval Augmented Generation (RAG) quality metrics help you identify problem patterns, conduct root cause analysis, and fine-tune your RAG-powered solutions.
About Knowledge/RAG Quality Data and Metrics: Learn about the metrics used to score retrievals. This feature collects data at run time and calculates quality scores for RAG-powered knowledge retrievals.
Implement Data Governance Permissions for Knowledge/RAG Quality Metrics and Data Objects: Knowledge/RAG Quality Metrics and Data can contain sensitive information, such as personally identifying information (PII), that you want to redact or hide from unauthorized users. Use Data Cloud Data Governance to protect sensitive data by setting up dynamic data masking and field-specific data exclusion policies. Grant access to the impacted Data 360 objects according to an access policy.
Set Up Knowledge/RAG Quality Data and Metrics: Turn on Knowledge/RAG Quality Data and Metrics to collect and store, in Data 360, quality scores and other data about knowledge retrievals. For orgs with only one dataspace, it’s turned on by default when Agentforce is turned on.
Agent Optimization: Use Agent Optimization to inspect agent sessions from the initial user request to the agent’s resolution.
About Agent Optimization: Agent Optimization provides tools to dive deeper into unresolved interactions, identify knowledge gaps, and analyze agent sessions using the Session Tracing Data Model. Agent Optimization is a key component of Agentforce Observability, designed to help you understand how your AI agents perform in real-world scenarios. By analyzing user interactions and agent responses, you can identify areas for improvement and take action to enhance agent effectiveness.
Use Intents in Agent Optimization: Work with intents in Agent Optimization to understand user requests, analyze interaction patterns, and improve your agent's performance.
Custom Scorer Prompt Examples: Use basic and advanced prompt examples to create custom scorers for common evaluation patterns.
Scorers and Custom Scorers: Scorers evaluate agent sessions for quality and business insights. You can score sessions with prompt templates that run automatically, by API, or manually in Agentforce Studio. Use out-of-the-box scorers, edit existing scorers, or create custom scorers with predefined or open-ended outputs.
Set Up Agent Optimization: To use Agent Optimization, turn it on in Einstein Setup. Analyze agent sessions, identify knowledge gaps, and explore unresolved interactions based on Agentforce Session Tracing data.
Troubleshoot Agent Optimization: Resolve common issues with Agent Optimization, including setup problems, data display issues, and performance concerns.
Analyze Data with Agent Optimization: Use Agent Optimization to drill into sessions, analyze quality scores, and identify patterns to improve your agent's performance.
Einstein Platform Reference: Review Einstein generative AI features, considerations, and implementation details.
Define Your AI Strategy: Bring an AI strategy to your company. Identify a vision, establish AI governance, define AI use cases, and develop your AI backlog.
Plan and Implement Your AI Project: Now that you’ve identified your first AI project, it’s time to start preparing for a successful project launch.
Prepare Your Data for AI: Data preparation can make or break your AI project. Learn how to define your data requirements, create a data inventory, and evaluate the data for AI readiness.
AI Project Success: To get the most out of your AI project, plan for AI project success. Every project is different, so choose the subagents that best suit your needs.
Learn How AI Can Transform Your Business: Where do you start with AI? Begin with inspiration for AI use cases and supplement those ideas with Trailhead resources. Here’s some ways you can get inspired about use cases for AI. You can also use the Business Process Exercise to identify areas in your processes where AI can help make your team more productive.
Geo-Aware LLM Request Routing on the Einstein Generative AI Platform: The Einstein generative AI platform routes large language model (LLM) requests to servers that are closest to where your Einstein generative AI platform instance is located. Geo-aware routing includes LLM API requests made by Einstein generative AI features that use the Einstein generative AI platform.
Geo-Aware Routing for Anthropic: LLM requests for Anthropic on AWS Bedrock are routed to the nearest AWS data center where a model endpoint is available.
Geo-Aware Routing for OpenAI and Azure OpenAI: A geo-aware model, such as GPT 4.1, uses OpenAI or Azure OpenAI to service LLM requests. The same underlying model is used by OpenAI and Azure OpenAI.
Agent Platform Tracing: This tracing capability provides comprehensive visibility into the actions and performance of Agentforce Agents, capturing crucial telemetry data across various services. By leveraging Data 360, users can unify this trace data with session information to create detailed reports. These reports are invaluable for monitoring Key Performance Indicators (KPIs), identifying performance bottlenecks, and quickly pinpointing the root cause of issues, ultimately enhancing the reliability and efficiency of Agent implementations.
Einstein Data Prism: Einstein Data Prism is a grounding solution for generative AI applications within Salesforce, improving the accuracy of AI solutions that use its grounding capabilities. With Einstein Data Prism, automatically ground your large language models (LLMs) so you can gain more accurate and relevant responses to utterances. Einstein Data Prism is automatically enabled in integrated apps, such as Einstein Segment Creation and Agentforce.
Metadata Studio: Use the Metadata Studio to enhance your generative AI results across Salesforce apps that are integrated with Data Prism, such as Einstein Segment Creation. With the Metadata Studio, review and validate all your objects and fields, add missing descriptions, and enhance AI-generated descriptions to gain an accurate semantic representation when querying natural language prompts. If needed, you can also exclude irrelevant objects in the data source so that they aren’t sent to the Large Language Models (LLMs) or returned when grounding.
Exclude Entities in the Metadata Studio: Use the exclude function when you want to exclude fields or objects that are no longer in use or are no longer relevant to agent use cases. If fields or objects are excluded, they're not used by Data Prism for grounding purposes. Exclusion only affects solutions that use Data Prism’s API.
Validate Metadata Entities in the Metadata Studio: To improve your Einstein results in Data Prism integrated apps, validate an object or field label, by adding or editing an entity description. Standard objects and fields with a description are automatically imported from Salesforce documentation and they’re marked as validated. You can still edit them if you have additional information that’s relevant.
Navigate the Metadata Studio: Explore the different sections of the Metadata Studio to better understand how to get the best results when validating your metadata.
Retrieval Augmented Generation: Use Retrieval Augmented Generation (RAG) in Data 360 to improve large language model (LLM) responses for agents and users. RAG is a framework for grounding prompts with relevant, accurate knowledge from unstructured data sources.
About Retrieval Augmented Generation: Learn how Retrieval Augmented Generation (RAG) in Data 360 enhances your Agentforce and Einstein generative AI solutions with knowledge.
Example: Agentic RAG with Advanced Data 360 Setup: This example walks you through the end-to-end steps to implement an agent that's grounded with knowledge using Retrieval Augmented Generation (RAG) in Data 360.
Agentforce Session Tracing: Agentforce session tracing captures detailed agent interaction data in a unified data model on Data 360. Build dashboards, reports, and queries on this data to derive insights into agent behavior and trends. Take data-driven action while testing, debugging, and monitoring agents in sandbox or production environments.
About Agentforce Session Tracing: Agentforce session tracing captures details on agent interactions with users from start to finish. Agentforce session tracing ingests data from different sources (starting with reasoning engine logs) and combines everything under a session ID. The log provides detailed information into what happened during agent sessions: turn-by-turn interactions, reasoning engine executions, actions, prompt and gateway inputs/outputs, error messages, and final responses.
Data Model for Agentforce Session Tracing: Use the data model in Data 360 to build reports, dashboards, and queries that provide insights into agent behavior. The system streams data to data lake objects (DLOs) in Data 360 within minutes. Access the data via the standard data model objects (DMOs), which are mapped to the DLOs.
Set Up Agentforce Session Tracing: Turn on Agentforce Session Tracing to collect and store, in Data 360, detailed data about agent sessions. By default, Agentforce Session Tracing is off. For orgs with only one dataspace, it’s turned on by default when Agentforce is turned on.
Einstein Trust Layer: Designed for Trust: The Einstein Trust Layer is a collection of features, processes, and policies designed to safeguard data privacy, enhance AI accuracy, and promote responsible use of AI across the Salesforce ecosystem.
Einstein Trust Layer: Response Journey: When the generated response is returned from the large language model, the Einstein Trust Layer applies certain policies and processes to make sure the response is safe and useful.
Build Trust in AI Responses with Citations: Citations link AI-generated responses to the grounding sources that are relevant to the response. Use citations to see what information the large language model (LLM) used to generate the response, and to verify the validity of the source data. Citations help you identify potential inaccuracies or hallucinations in the generated responses, increasing your confidence in using AI tools.
Configure Citation Mode for Custom Actions: With citations mode, custom actions that are created by using prompt templates can be configured to show source URLs in the generated response.
Turn On Prompt Injection Detection for Prompts (Beta): Turn on prompt injection detection in Einstein Trust Layer Setup to monitor content that attempts to make the large language model do something that it isn’t designed to do. These settings are applied at the Salesforce org level.
Turn On Toxicity Detection For Prompts (Beta): Turn on toxicity detection in prompts so you can monitor the prompts that have toxic language and modify them accordingly. These settings are applied to your Salesforce org.
Content Safety and Security: Einstein Trust Layer includes a set of policies to help detect potentially harmful content and malicious attacks that attempt to compromise the safety and security of AI applications.
Einstein Trust Layer Region and Language Support: The Einstein Trust Layer detects and masks sensitive data such as personally identifiable information (PII) for multiple regions and languages.
Einstein Trust Layer: Trusting generative AI with your company secrets and your customer data while balancing the benefits of generative AI is still a challenge today. The Einstein Trust Layer is a secure AI architecture, built into the Salesforce platform. It’s a set of agreements, security technology, and data and privacy controls used to keep your company safe while you explore generative AI solutions.
Einstein Trust Layer Limits: Einstein Trust Layer is supported in the sandbox staging environments with some limits.
Einstein Generative AI & Trust: It’s important that your data stays safe while you innovate with new technology. Einstein generative AI keeps trust as its #1 value, and we strive to make sure your data is secure while also creating experiences that are accurate and safe.
Prompt Injection Detection (Beta): Prompt Injection, sometimes referred to as “jailbreak,” is a large language model (LLM) vulnerability. Prompt injections are attempts to make the LLM do something that it isn’t designed to do. Hackers can create prompts that attempt to override the system policies or manipulate the LLM into doing something unintended.
Set up Einstein Trust Layer: Configure the Einstein Trust Layer to align with your organizational privacy and security policies and requirements.
Show Sources When Using Answer Questions with Knowledge Standard Action: The Answer Questions with Knowledge standard action currently supports prompt-based citations that display the source URLs at the end of the response. To include in the generated response, you must enable the option to show sources for the agent.
Toxicity Detection: Einstein Trust Layer uses machine learning (ML) models to identify and flag toxic content in prompts and responses.
Agentforce Conversations Billable Usage Types: Usage of certain Agentforce features impact credit consumption against subtypes of the Agentforce usage type on the Conversations consumption card. It consists of three subtypes: ASA Messaging, Sales Coach, and SDR.
AI Agent Generative AI Usage Data Model: Use the AI Agent Generative AI Usage data model object (DMO) to analyze Agentforce and generative AI consumption patterns in Data 360.
Einstein Requests Billable Usage Types: Usage of certain generative AI features impacts credit consumption against subtypes of the Einstein Requests billable usage type. Usage is calculated based on the number of calls to the LLM gateway. Einstein Request usage types apply to embedded generative AI features such as Service Replies that involve direct requests to the LLM for specific tasks.
Metering for Agentforce and Generative AI Usage: Generative AI usage for Agentforce features is typically metered and consumes credits.In some cases, when required conditions are met, usage can be unmetered and doesn’t consume credits.
Configure User Permissions for Unmetered Licenses: To enable unmetered AI usage, first assign the appropriate permission set license to the user where available. Create a permission set for the license, and then finally assign the permission to the user.
How Agentforce Help Agent Is Billed: Agentforce Help Agent supports resolution-based, login-based, and member-based billing, depending on your license and how you deploy the agent.
Billing Outcome Details: This chart shows the specific reasons behind each billing outcome with percentages, plus drill-down into individual conversations and turn counts. Filter by "Not Resolved" to uncover knowledge gaps or spot user drop-off points.
Billing Outcomes and Reasons by Channel: This chart breaks down billing outcomes and reasons by channel to compare agent performance across them. Use it to spot channel-specific escalation spikes or low adoption.
Billing Outcomes Over Time: This chart tracks resolved versus unresolved outcomes and volume day by day to reveal trends rather than one-off snapshots. Rising unresolved rates or sudden volume drops signal issues worth investigating.
Chart Guidance and Actions: Use the billing resolution analytics charts to spot channel-specific escalation spikes, resolution trends, and knowledge gaps. Use this guidance to spot channel-specific escalation spikes, resolution trends, and knowledge gaps. Then take the recommended actions, such as reviewing escalation flows, updating the grounded knowledge, or refining the welcome greeting.
Customize Help Agent Default Settings: While the Help Agent comes with default configurations designed to support common service workflows out of the box, you can tailor these settings to better align with your specific business requirements. To customize out-of-the-box solutions or create your own, use Agentforce Builder and agent assets, including subagents, actions, filters, and variables.
Edit a Help Agent with Quick Service Agent Configuration: Each time you edit and redeploy an agent, a new version is created to track changes over time. Whenever possible, use Quick Service Agent Configuration to manage your Help Agent, such as when updating knowledge grounding sources.
Understanding Escalations from Help Agent to Service Rep: Customer conversations with the Help Agent can be escalated to a human service rep. By default, escalation is triggered when a customer explicitly requests to speak with a human or service rep.
Configure a Feedback Prompt After Task Completion: Add a feedback step immediately after your agent completes a task or answers a question, so it captures the outcome before the conversation ends.
Manage Your Help Agent: Manage your Help Agent to finish where you last left off or to improve its accuracy and effectiveness. For example, you can edit or delete an existing agent, view details for each agent version, and deactivate or activate an agent.
Manage User Access to Build a Help Agent: To build a Help Agent, you must be assigned the Help Agent Configuration permission set license and the corresponding Quick Service Agent Configuration permission set. To turn on Data Cloud, which is a prerequisite for building a Help Agent, you must be assigned the Data Cloud permission set license and the corresponding Data Cloud Architect permission set. The user who turns on Help Agent is automatically assigned both permission set licenses and permission sets.
Set Up a Help Agent: Help Agents provide 24/7 self-service support to your customers across web chat, help portals, and voice channels, grounded in your knowledge base. Use the Help Agent template to quickly create and configure an agent that resolves customer issues and seamlessly escalates to service reps when needed.
Test and Go Live with Your Agent: After you build or update a Help Agent, run tests to evaluate the agent’s performance and accuracy. For example, you can compare the agent's actual response to the expected one. In Testing Center (Beta) in Agentforce Studio, you can build test suites and run multiple test scenarios simultaneously. A test suite run can catch intermittent bugs, like skipped actions and even things missed while previewing the agent.
Turn On Help Agent Analytics: To enable users to view Help Agent performance and resolution metrics, turn on Help Agent Analytics. Help Agent Analytics covers resolution events from the point when Help Agent Analytics is turned on. Historical records aren't backfilled.
Agent Versions and Channels: When you create a Help Agent, Salesforce tracks your changes using versions. Understanding what is and isn't version-controlled helps you make changes safely and plan deployments effectively.
View Help Agent Analytics: Monitor key performance metrics, resolution rates, and flex credit consumption for your deployed Help Agents to track support effectiveness and customer satisfaction.
Migrate from Agentforce (Default) to Agentforce Employee Agent: Discover the key differences between Agentforce (Default) and Agentforce Employee agent, and learn how to seamlessly migrate your default agents. The migration flow creates a new Employee agent with the same subagents and actions, variables and settings as your default agent. You can fine-tune your Agentforce (Default) agent into multiple employee agent personas, each specializing in different use cases, with improved access control and channel integrations.
Monitor Your Voice-Enabled Agent: To boost service quality, supervisors can monitor the voice-enabled agents on active calls by using Omni Supervisor. They can also monitor agent performance over time by using agent analytics reports and dashboards.
Federated Search in Experience Cloud Sites: Finding content is key for your customers. But when content is brought in using CMS Connect (JSON), it’s only displayed in Salesforce, not hosted there. Standard searches can’t find that content unless you implement federated search to include the source site.
Add a Prompt Template as an Action: Add a prompt template as an action within another prompt template to reuse logic and incorporate its output into your prompt. Invoke one prompt template from another to build consistent, modular prompt workflows.
Add a Field Generation Prompt Template to a Lightning Record Page: Generative AI-enabled fields powered by prompt templates can help users complete their work faster. With a click of a button, your users can quickly populate a record field with useful information, such as a summary of open cases for an account, to facilitate more productive conversations with your customers.
Add Record Merge Fields to a Field Generation Prompt Template: Save time with templates that use specific information from your Salesforce objects to create case summaries, and more, at scale. Use merge fields in a field generation prompt template to have an LLM write content by using up-to-the-minute data.
Add Record Merge Fields to a Flex Prompt Template: Save time with templates that use specific information from your Salesforce objects to create unique and detailed product descriptions, custom text messages, and more, at scale. Use merge fields in a flex template to have an LLM write content based on up-to-the-minute data.
Add Record Merge Fields to a Sales Email Prompt Template: Merge fields are critical to scaling a sales email prompt template for multiple customers and products. They connect your prompt templates to Salesforce record fields, such as from sales or service records. In a prompt template, a merge field acts like a placeholder. After you run the template, the LLM replaces the merge field with your actual data.
Add a Sales Email Prompt Template to Einstein Sales Emails: Get ready to add sales email prompt templates from Prompt Builder to your sales team’s toolbox. Turn on Sales Emails in Setup, and then assign a permission set to your sales reps.
Use AI Assist in Prompt Builder (Beta): Use AI Assist to create a structured, Jinja-based prompt template from a natural language description. AI Assist is powered by an Agentforce agent that pulls in relevant fields from your org and structures prompts using role, task, and audience formatting.
Create an Answer Questions with Knowledge Prompt Template: Enable agents to answer questions using the Answer Questions with Knowledge prompt template in Prompt Builder. Specify what information Einstein includes, or adjust the voice and tone so that answers fit your business needs.
Build and Manage Lightning Types for Prompt Templates: Use Lightning Types to define and manage structured output schemas for prompt templates. Lightning types are reusable definitions that Prompt Builder references at run time to enforce the response structure.
Create a Prompt Template: Select the type of prompt template to create based on your use case. Based on the template type, define the resources available for the template.
Choose the Right AI Model for Your Prompt Template: Select and test AI models to find the best fit for your specific use case. The right model choice significantly impacts response quality, cost, and performance.
Pass Data Between Actions in Prompt Builder: Pass data between actions in Prompt Builder to create dynamic and flexible prompt workflows. Use the output of one action as input to another to generate more contextual and relevant responses.
Change Set Considerations for Prompt Templates: Before you use change sets to deploy a prompt template, understand the limits and behaviors that are related to component dependencies and deployment.
Salesforce CLI Considerations for Prompt Templates: Understand the limits and behaviors that are related to metadata component dependencies and deployment when using Salesforce CLI with Prompt Templates.
Deployment Considerations for Prompt Templates: Include a prompt template in managed, unmanaged, or unlocked packages. Before you deploy a package that contains a prompt template, in changesets or on the CLI understand the limitations and behaviors.
Packaging Considerations for Prompt Templates: Include a prompt template in managed, unmanaged, or unlocked packages. Before you create, update, or deploy a package that contains a prompt template, understand the limitations and behaviors of packages.
Create a Field Generation Prompt Template: A Field Generation prompt template brings generative AI-assisted workflows to fields within Lightning record pages. Enable your users to populate a field with a summary or description created by an LLM to facilitate more productive conversations with your customers.
Create a Flex Prompt Template: For any use case that isn’t covered by the other template types, a flex prompt template type is your solution. Craft a flex prompt template that uses AI to generate content for your unique business cases.
Create a Sales Email Prompt Template: Connect with customers using emails that leverage your customer relationship history and insights. To generate personalized, targeted emails at scale, design a sales email prompt template. Because you can reuse your template for different products and customers, getting an LLM to write your emails saves you time and makes email workflows easier.
Enable Prompt Builder: Enable Einstein in your organization and assign users the appropriate permission set to use Prompt Builder.
Enable Open Response Language for Prompt Templates: Use the Automatic toggle to allow the LLM to interpret language directly from your prompt instructions instead of using system-managed language selection.
Enabling Resolution to Review a Prompt Template: When you enable resolution, you can review the resolved prompt that’s generated from your prompt template. Through a process called prompt resolution, Prompt Builder replaces each merge field in the prompt template with real CRM data related to the records that you selected.
Evaluate the Quality of a Response: When you preview a generated response, evaluate how well the response meets the goals of the prompt. Make sure that the response is factually accurate and doesn’t contain harmful content or bias.
Example Prompt Template Library: This library of example prompt templates gives you a starting point for creating your own custom prompt templates.
Limit Field Generation Responses to Business-Related Activity: By default, field generation follow-up refinement requests aren't validated to ensure relevance to business requests. This means users can make unrelated, non-business requests within the field generation panel, such as "Write a haiku." Use the Validate Field Generation Refinements setting and prompt template to make sure that user follow-up requests stay focused on business use cases and protect your Einstein credits from being spent on unrelated tasks.
Get to Know Prompt Builder: Prompt Builder features a guided, visual layout that takes you through each stage of your prompt’s lifecycle, from creation, to testing, and activation.
Give Users Access to a Prompt Template: Prepare your users to harness the power of Einstein generative AI. Manage who can use prompt templates with permission sets. Add sales email prompt templates to Einstein for Sales, and add field generation prompt templates to dynamic fields in pages built with Lightning App Builder.
Grounding with Apex Merge Fields: You can use an Apex merge field in a prompt template to return data from a SOQL query or an external API. Apex is also effective if you want to generate well-formatted JSON or do programmatic data filtering.
Grounding with Data Graphs: Unify multiple data sources into a centralized dataset to access a 360-degree view of customer data.
Grounding with File Inputs: Ground generative AI prompt templates with files so the AI model can use documents, spreadsheets, presentations, images, and email text in the resolved prompt.
Ground with Flow Merge Fields: Flows elevate your prompt templates beyond grounding with record fields. Use Flow Builder to add complex logic to a prompt template. With flows, you can also access unified Data 360 objects. By referencing Data 360 data, you can use real-time information that was consolidated across multiple data sources.
Ground with Record Merge Fields: Record merge fields connect your prompt templates to Salesforce object fields. The prompt template type and the objects that you associated with your prompt template determine which merge fields are available.
Ground with Predictive AI: Ground your prompt templates with relevant knowledge and actionable intelligence from predictive models. Augment your LLM prompts with predictions, top predictors, and prescriptions.
Ground with Knowledge Using Retrieval Augmented Generation: Retrieval augmented generation (RAG) in Data 360 improves the quality, accuracy, and relevance of your LLM-generated responses. Enhance your prompt templates using retrievers with Einstein Search. Add relevant knowledge from unstructured text sources into your LLM prompts. Retrieve knowledge from articles, emails, chat transcripts, and other sources.
About Retrievers in Prompt Templates: Use retrievers in prompt templates to search for and return relevant information from knowledge that's indexed in Data 360. Retrievers augment prompt templates by providing relevant, specialized grounding information for prompts. Retrievers are created in AI Models (formerly Einstein Studio).
How Users Work with a Prompt Template: As a Salesforce administrator, you configure prompt templates to make tasks easier for your users. This section explains how your users can apply these prompt templates to write emails, generate case summaries, complete text form fields, and more.
Prompt Builder: Use Prompt Builder to build, customize, and manage AI prompt templates that incorporate your CRM data to power generative AI features and Agentforce agents.
Work with Large Prompts: When a prompt is too large for the model to use, a summary is generated automatically in the Resolution panel using OpenAI models . You can decide to use the summarized prompt or create a smaller prompt.
Prompt Builder Limitations: Learn about the functional limitations of prompt templates that affect certain behaviors and outcomes.
Supported Languages in Prompt Template Responses: A prompt template can use any language that Prompt Builder offers. Supported languages are optimized for business-appropriate phrasing and are integrated with the Einstein Trust Layer to use its data masking and toxicity scoring features.
Manage Changes to Object-Based Types: You create object-based custom Lightning types in Prompt Builder to structure prompt template output. When you edit these types, you can’t modify certain field properties to avoid disrupting integrations. Modifications such as renaming fields, changing data types, deleting fields, or making optional fields required can cause integrations and references to fail.
Manage Prompt Templates: Fine-tune your prompt template in Prompt Builder by using versions, activation and deactivation, changing LLM configurations, and more.
Optimize and Refine Prompt Templates: Iteratively improve your prompt templates to achieve consistent, high-quality results. Great prompt templates are rarely written perfectly on the first attempt.
Organize Complex Prompt Templates: Structure prompts with multiple data sources and merge fields for clarity and consistency. Well-organized prompts reduce errors, speed up troubleshooting, and produce more reliable AI output.
Override a Standard Prompt Template: You can override some standard prompt templates to create one or more custom versions. Only one version can be active at a time. When you activate a standard prompt template override, it takes the place of the standard version.
Evaluate Prompts with Performance Insights: Analyze model generation metrics and related insights directly in Prompt Builder to understand performance, token usage, and masked data when testing your prompt templates.
Customize Retriever Output with Prefilters: Get more relevant search results using prefilters in your prompt templates. Prefilters define the conditions that limit the scope of a retriever’s search so that only the most relevant information is returned. For example, you can use prefilters to retrieve content associated with only a particular country or product. The filter value, such as the country code or product Id, is specified in the prompt template at run time.
Preview a Resolved Prompt: In the Resolved Prompt section of your workspace, you can review a prompt that’s grounded with real Salesforce record data.
Preview a Generated Response: In the Response section of your workspace, you can review a response that’s grounded with real Salesforce record data.
Prompt Template Output Languages: You can configure the languages that a prompt template can use in its output. Languages that you select appear in the Preview Language picklist in the Prompt Builder workspace.
Create Prompt Templates with AI Assist (Beta): Use AI Assist to create a Jinja-based prompt template from a natural language description. Review the draft in the canvas and customize it with additional instructions before accepting.
Remove a Template Reference in an Agent: Remove dependencies on a prompt template so you can safely delete or modify the template without breaking other components.
Remove a Template Reference in a Flow: Remove dependencies on a prompt template so you can safely delete or modify the template without breaking other components.
Add Citations to Prompt Template Responses: Add citations to your prompt templates so that your users can compare and verify the generated information with the source information. Sources include content used by agent actions, knowledge articles, and web pages.
Add an Ensemble Retriever to a Prompt Template: Use ensemble retrievers in prompt templates to ground your LLM prompts with relevant knowledge. An ensemble retriever is a collection of individual retrievers. When you run an ensemble retriever, it executes the individual retrievers, combines their search results into a single list, reranks the list according to relevance to the search request, and returns just the most relevant information to the prompt or agent. Ensemble retrievers conduct searches from multiple sources in parallel.
Add an Individual Retriever to a Prompt Template: Use individual retrievers in prompt templates to ground your LLM prompts with relevant knowledge. Individual retrievers include retrievers created in AI Models (formerly Einstein Studio) and legacy default retrievers created previously in Data 360.
Add a Web Retriever to a Prompt Template: Use web retrievers in prompt templates to ground your LLM prompts with relevant knowledge from web sites. Web retrievers are created when a data library that uses web search is created in Agentforce Data Library.
Review, Revise, and Repeat: Ensure That a Prompt Template Is Effective: Generative AI uses LLMs to generate output. Because LLMs are trained by other organizations, it’s common to first receive an LLM response that isn’t quite right, even if the prompt seems clear. If the output doesn’t meet your expectations, you can generate a new output by revising your prompt template. The generated output from the previous attempt isn’t saved, and the new output replaces it. With Prompt Builder, it’s easy to edit your prompt template until you’re completely satisfied with a response.
Revise a Prompt Template: Perfecting a prompt template is an iterative process. You revise the prompt template, regenerate a resolved prompt and response, and then make further revisions based on that updated response. Every time you run a prompt template in Prompt Builder, you create a unique call to the LLM. That means that resolving the same prompt can produce different responses. Remember to review every response for accuracy, even if you’ve run the template before.
Prompt Template Run-time Execution: Invoking a prompt template at run time sets off a chain of events. The run-time life cycle of a prompt template begins with loading the prompt template with input and concludes with forwarding the LLM-generated response to the calling agent or app.
Configure Language Selection for Prompt Templates: Manage which languages your prompt template can generate in and how the language is selected. Configure system-managed language selection based on user and org settings, or enable instruction-driven language selection where the LLM interprets response language directly from prompt instructions.
Standard Prompt Templates: Use standard prompt templates as starting points for typical use cases. You can customize a standard prompt template by copying it. You can override some standard prompt templates to create versions that work in place of the standard version.
Prompt Template Types: Learn about the prompt template types available in Prompt Builder.
Ingredients of a Prompt Template: When you start to build your own prompt templates, it can feel intimidating. Like all new skills, designing great prompts requires some experimentation and practice. To guide your prompt design, it’s helpful to start with the key components of every effective prompt template. Like a good recipe, these “ingredients” work together to help large language models (LLMs) generate accurate and consistent responses.
Prompt Template References: Identify where a prompt template is used across your org. When you view references for a template, you can see the flows, agents, and other components that depend on that template.
Field Generation Prompt Templates in Action: With generative AI-enabled fields, you can get real-time data from across your org without leaving a Lightning page. Generative AI-enabled fields use field generation prompt templates to provide dynamic summaries, descriptions, and more.
Flex Prompt Templates in Action: Flex prompt templates are specific to your use cases, so call them wherever you need them in your org. Build custom UI or use REST API to call them with an invocable action.
Sales Email Prompt Templates in Action: Einstein generative AI helps you draft a personalized email. Your org’s custom sales email prompt templates automatically appear as options in the Lightning Experience email composer. Before you send an Einstein Sales Email, review the email draft and edit it, if needed.
Understand the Block Editor: The Block Editor is a visual, block-based canvas for building prompt templates. Add resources, format instructions, apply conditional logic, and rearrange text without writing HTML.
Use Conditional Logic in Prompt Templates: Create dynamic, context-aware prompt templates using conditional logic in the Block Editor. Add if, else if, and else statements to tailor AI responses based on input variables and grounded data.
Use Example Prompt Templates: To get started with Prompt Builder, customize one of the example prompt templates from the library.
Use Prompt Templates as Actions: Add a prompt template as an action within another prompt template to reuse logic and standardize how responses are generated. Invoke one prompt template from another to build more consistent, modular prompt workflows.
Use Structured Outputs to Format Model Responses: Structured Output in Prompt Builder ensures models return responses in a consistent JSON or HTML format, improving reliability for workflows, integrations, and UI rendering.
View Performance Insights in Prompt Builder: Evaluate key generation metrics like response time, token usage, and masked values directly in Prompt Builder to understand how your prompt runs.
View Template References: Use Template References to identify where a prompt template is used and safely remove those dependencies before making changes.
Standard Dashboard for Einstein Lead Scoring: The standard dashboard for Einstein Lead Scoring helps you track and manage lead conversion rates. All Sales Cloud Einstein users have access to standard dashboards.
Agentforce Coworker: Powered by Data 360, Agentforce Coworker is your AI-powered agent, connecting all your work tools into one conversational interface. Instead of juggling several different apps and pages, you can use Agentforce Coworker to find information, complete tasks, and work with your team and your customers, all in one place.
Select Which Objects to Always Search in Global Search: For new and inactive users, Einstein sometimes doesn’t have enough data to accurately determine which objects to search. For these users, search results can exclude important objects. Select objects to always search for a user profile to ensure that important objects are always included. You need a global search configuration to define objects to always search.
Explore Einstein Search: Einstein Search gives the global search bar more power and intelligence from the first click. Work faster, complete tasks directly from the search results, and get results that are most relevant to you. With personalization, actionable instant results, and natural language search, Einstein Search can help improve your productivity.
AI-Generated Answer Card Examples: Search Answers supports multiple channels such as the global search bar, service rep channels, and end-user channels with Salesforce Cloud Experience sites. See examples of AI-generated answer cards across supported channels.
Use Case: Make Your Knowledge Articles Available in Einstein AI-Generated Search Answers: Provide relevant search results by making your knowledge articles available in Einstein AI-Generated Search Answers. Use the Service Cloud data kit to map your knowledge articles to a Data Cloud DMO. By selecting the DMO as a source object for Search Answers, service agents can quickly find answers to customer questions.
Filter Out Your Knowledge Search: Narrow the focus of your knowledge search and find articles that are the most helpful to you. From the global search bar, you can filter out your search criteria to knowledge.
Natural Language Search Examples: Start a natural language search using one of the supported objects. You can use a combination of words and phrases. The table shows examples for each object. After you define a word, enter your search terms, and then press Enter (Windows) or Return (Mac) to see your search results.
Use Natural Language Search: Natural language search (NLS) gives you quick access to lists of records so that you can find the information that you need. Enter everyday words and phrases, such as my open opportunities this month or cases closed last week in the global search bar.
Enhance Natural Language Search: Make natural language search more personalized and useful. You can receive results based on your location and rename objects to names more relevant to your business, and rank the opportunities and leads that are most important to you. Also, you can explore how picklists and operators break down your search.
Limitations for Non-English Terms: Natural language search offers full support for English language queries. It can also understand queries using non-English languages, but there are some limitations. If a non-English query isn’t supported, Einstein Search does a keyword search instead.
Get Relevant Search Answers: Resolve issues quickly and deflect cases with Einstein Search Answers, AI-Generated Search Answers, and AI-Generated Search Answers from Data Cloud sources. Enter a question or phrase in the global search bar, the Knowledge sidebar on the case console, or from Experience Cloud sites. Search Answers extracts the most relevant response or provides an AI-generated response from your knowledge or Data Cloud data model object (DMO) sources. Search Answers supports only English queries and searches only English language articles and DMO fields.
Search Answers Requirements and Considerations: Standard Search Answers, AI-Generated Search Answers, and AI-Generated Search Answers from Data Cloud sources have different requirements and capabilities.
Search Analytics Reference: Learn about the data model objects (DMOs) and calculated insights (CIs) that Search Analytics uses.
Einstein Search Best Practices: Whether you search for records or receive a recommended result, sometimes search provides you with a result that’s not relevant to your term. Here are some tips to help improve your search experience.
Search in Salesforce Classic: Search for the right records to get your work done. If enabled, use global search, at the top of every page, to search across Salesforce.
Lookup Search Glossary: Use this cheat sheet to view what Lookup features are available in Salesforce Classic.
Use Salesforce Search: Make the most out of your search with the right tools. Easily search for records, integrate lookup searches, and further refine your results in Classic.
Troubleshoot Einstein Search Cases: Einstein Search varies for each user. Here are some common scenarios that occur when using Einstein Search.
Customize Search in Experience Cloud Sites: Your partners, employees, and customers in Experience Cloud sites have different needs when searching for information. Sites also address specific use cases, including portals, help forums, and support communities. Create a search configuration with the needs of a site in mind. Choose which objects and fields are available to search, and filter out content they don’t need.
Publish the OpenSearch Description: To transform the Solr default search results from JSON to XML Atom, create and publish an OpenSearch description file. This XML file defines the search query URL and supported parameters.
Edit and Test the Solr Query: Add parameters that reference the Atom XSL file to the Solr query. Then test your changes to make sure that results are appearing in XML Atom format.
Configure a Solr Server to Support OpenSearch: To show external search results in Salesforce, set up a Solr server. Configure the Solr server to support OpenSearch and Atom XML format. The OpenSearch standard is the basis for Salesforce Federated Search, which displays external search results in Salesforce. After you configure the Solr server, define the external data source providing search results.
Create the Atom XSL File: To transform the Solr default search results from JSON to XML Atom, you create an XSL file. Solr provides example files that you can customize for your server.
How Search Crowding Affects Search Results: Search crowding–having too many unhelpful search results–can overwhelm you with irrelevant records and slow down search. To avoid search crowding, the search engine limits the number of records it analyzes. Learn how to adjust your searches for the best results.
Use an External Provider Data Source for Federated Search: Let users search in your Salesforce org and access data from an external search provider. Salesforce supports external search providers that support the OpenSearch standard.
Disable an Object Search Index: Disable indexing for an object to stop making its data searchable and stop index credit consumption for that object.
Edit Your Search Layout: Take a look at how you can edit and customize your search layout to best fit your users needs.
See Instant Actionable Results: The moment you put your cursor in the global search bar, you see suggested searches and record previews based on your recent activity. Use these results to easily access records. You can also complete actions, such as editing a record and accessing related quick links.
Get Personalized Results: Personalization is an AI feature that reorders search results based on the objects and records you work with most. Each time you search, personalization learns more about your preferences, which improves your search results.
Recommended Results: Einstein Search shows you a recommended result when it finds a good match for your search terms. It places this result first along with actions and links to related records.
Configure Lookup Search Results in Salesforce Classic: After you’ve defined lookup fields, configure your search settings. Enable enhanced lookups and lookup auto-completion, and start customizing your lookup filter fields.
Enable Enhanced Lookups: Enable enhanced lookups in Classic so users can sort, filter, and page through their results. Enhanced lookups are available only for specific objects.
Enable Lookup Auto-Completion: Enable lookup auto-completion so users can select items from a dynamic list of matching, recently used records when editing a lookup field. It’s supported for account, contact, user, opportunity, and custom object lookups.
Specify Lookup Search Filter Fields: After enabling enhanced lookups, specify which fields users can use to filter lookup search results. If you don't specify any fields, your users can't use filters in enhanced lookup dialogs. Enhanced lookups are available only for specific objects.
Tips on Search Errors: Search sometimes experiences errors because of a network connection, complex searches, index problems, or other reasons. You can still use search when the system is down but with some limitations. Search is usually restored in a few hours, but it can take longer. Keep some guidelines in mind when using search with errors.
Search Analytics Calculated Insights: Search Analytics provides calculated insights (CIs) containing metrics about queries, articles, and interactions. Each CI is created with data model objects (DMOs). These tables contain the relationships between the CIs and DMOs.
Create a Case Closure Calculated Insight: Create a calculated insight (CI) with case-related metrics by using example SQL expressions. Use the CI to see how many interactions happen before the case is closed, to analyze case-related metrics, and to build dashboards.
Access the Search Analytics Dashboards: Search Analytics provides dashboards to analyze data about search in your Salesforce org. The dashboards contain graphs to view trends about queries and articles.
Give Users Access to Search Answers or AI-Generated Search Answers: Use permission sets to give service reps and Knowledge users access to Search Answers, AI-Generated Search Answers, or AI-Generated Search Answers from Data Cloud sources. Only users with the Search Answers permissions can get responses from Search Answers when they search in the global search bar or Knowledge sidebar.
Use Natural Language Search Filters (Pilot): Einstein Search can understand a natural language search in different ways. If Einstein doesn’t interpret your search as you intended, you can change the filters that Einstein applied to the search. After you select new values, the search results are updated in real time.
How Does Einstein Search Use My Data: Get answers to your questions about how Einstein Search uses your data for global models. Rest assured that nothing changes in terms of security, privacy, or data residency.
Enable Search Analytics: To use Search Analytics, you need a Data Cloud license and Data Cloud credits. To include user data in Search Analytics, you must first enable search analytics.
Enable Search Answers or AI-Generated Search Answers: Search Answers gets the most relevant snippets from your data sources, and AI-generated Search Answers uses knowledge articles and other sources to generate tailored responses to user questions. Search Answers supports only English queries and searches English-only knowledge articles or Data Cloud data model object (DMO) fields.
Manage Einstein Search for Knowledge Settings: Let administrators manage the settings of Einstein Search for Knowledge, so that service agents and their customers use Einstein Search for Knowledge features for easier access to knowledge articles related to their work.
Dive into Einstein Search for Knowledge: Artificial Intelligence (AI) powers Einstein Search for Knowledge to find the most relevant knowledge articles for service agents and their customers. Improve your filtering experience and use multi-language search support to narrow down your search results faster. Einstein Search for Knowledge remembers your filters and saves you time when you search for similar articles.
Get Insights with Search Analytics: Discover actionable insights with Search Analytics. Build dashboards to analyze your data. Use insights to make changes to the search engine, optimize content, and train users.
Billing Considerations for Search Analytics: Search Analytics gives you visibility into search performance through dashboards that show usage and behavior metrics. Search Analytics runs on Data 360 calculated insights, queries, and data prep, which impacts credit consumption in these usage types rather than incurring a flat fee.
Update the Date Filter in Search Analytics Calculated Insights: Calculated insights (CIs) power the data in the Search Analytics dashboards. Some Search Analytics CIs filter out older data so you can make decisions based on the most up-to-date information. To customize CIs and see data for specific data ranges, you can update the SQL expressions.
Understand the Search Analytics Dashboard: Explore the components in the Search Analytics Dashboard, including the data model objects (DMOs) and calculated insights (CIs) that power the data and what each chart or table shows.
Billing Considerations for Search Answers: AI-Generated Search Answers from Data 360 sources creates a hybrid search index for source data model objects (DMOs) and uses generative AI to respond to user questions. Use of AI-Genererated Search Answers impacts the consumption credits in these usage types.
Optimize Knowledge Articles for Search Answers: To improve search results with AI-generated Search Answers, review your source knowledge articles or Data Cloud data model object (DMO) text fields and optimize titles and body text.
Select Data Model Objects as Sources for Search Answers: Specify Data Cloud data model objects (DMOs) and fields to use as sources for AI-generated Search Answers. Use Lightning App Builder to create a Lightning record page for each DMO so that users can see the source records for the AI-generated summaries.
Manage Einstein Search Settings: Einstein Search is on by default for your entire Salesforce org and enhanced with features at no additional cost.
How Article Results Are Ranked: Agents get relevant results with the latest AI models in Einstein Search for Knowledge. Your models can learn complex relationships between your articles and provide you with the most relevant articles in your top results.
Apply Multi-Language Knowledge Search: Multinational companies have service centers around the globe, and their knowledge base reflects that geographic diversity. Service agents can get knowledge article results in different languages.
Configure the External Object: Now that the search information is identified and the data available, you can make it available to the Community User Profile.
Create the External Object: Enabling federated search is a two-phase process. First include the external site in the search engine scope, and then expose the results to your customers.
Enable the Object in Experience Cloud Sites: Now that you’ve added the connection and set all the permissions, expose the results to the customers in your community. To display the results, add them to the Global Search Results component of Experience Builder.
Explore the Search Index with Search Manager: See default searchable objects and fields in your Salesforce org. As you add custom objects and fields to your Salesforce org, your index changes to reflect which objects and fields are searchable. Your index is the single source of truth for what’s searchable in your org. You can manage the search index and apply field-level security to custom fields.
Federated Search Considerations: When setting up federated search, familiarize yourself with the nice-to-knows and limitations.
Set Up Federated Search: Do your users search for content stored outside of Salesforce? Federated search makes it easy to add external search engines (or connectors) to your org. Users look for information using Salesforce global search and see external results in a single search results page. Understand more about how federated search works, how to set it up, and how users see results.
How Federated Search Works: Users can search for data stored in repositories outside of Salesforce while remaining inside the Salesforce user interface. Get a high-level look at how Federated Search works.
Searchable Fields: Account Brand: Searchable fields for Account Brand Salesforce Classic include various standard and custom fields that are relevant to the account brand. These fields can depend on an organization's configuration and customization.
Searchable Fields: Account Relationship: Searchable fields for Account Relationship Salesforce Classic include various standard and custom fields that are relevant to account relationship. These fields can depend on an organization's configuration and customization.
Searchable Fields: Activities (Events and Tasks): Searchable fields for Activities Salesforce Classic include various standard and custom fields that are relevant to Activities. These fields can depend on an organization's configuration and customization.
Searchable Fields: Address: Searchable fields for Address Salesforce Classic include various standard and custom fields that are relevant to the address object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Knowledge Article: Searchable fields for Knowledge Article Salesforce Classic include various standard and custom fields that are relevant to the knowledge article. These fields can depend on an organization's configuration and customization. Sidebar search and advanced search aren’t designed to find articles. To find an article, use global search or the search tools in the sidebar on the Articles tab.
Searchable Fields: Asset: Searchable fields for Asset Salesforce Classic include various standard and custom fields that are relevant to the asset object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Asset Relationship: Searchable fields for Asset Relationship Salesforce Classic include various standard and custom fields that are relevant to the asset relationship object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Asset Warranty: Searchable fields for Asset Warranty Salesforce Classic include various standard and custom fields that are relevant to the asset warranty. These fields can depend on an organization's configuration and customization.
Searchable Fields: Attachment: Searchable fields for Attachment Salesforce Classic include various standard fields that are relevant to the attachment. These fields can depend on an organization's configuration and customization.
Searchable Fields: Authorization Form: Searchable fields for Authorization Form Salesforce Classic include various standard and custom fields that are relevant to the authorization form. These fields can depend on an organization's configuration and customization.
Searchable Fields: Authorization Form Consent: Searchable fields for Authorization Form Consent Salesforce Classic include various standard and custom fields that are relevant to authorization form consent. These fields can depend on an organization's configuration and customization.
Searchable Fields: Authorization Form Data Use: Searchable fields for Authorization Form Data Use Salesforce Classic include various standard and custom fields that are relevant to authorization form data use. These fields can depend on an organization's configuration and customization.
Searchable Fields: Authorization Form Text: Searchable fields for Authorization Form Text Salesforce Classic include various standard and custom fields that are relevant to authorization form text. These fields can depend on an organization's configuration and customization.
Searchable Fields: Award: Searchable fields for Award Salesforce Classic include various standard and custom fields that are relevant to the award. These fields can depend on an organization's configuration and customization.
Searchable Fields: Thanks Badge: Searchable fields for Thanks Badge Salesforce Classic include various standard fields that are relevant to the thanks badge. These fields can depend on an organization's configuration and customization.
Searchable Fields: Business Account: Searchable fields for Business Account Salesforce Classic include various standard and custom fields that are relevant to the business account. These fields can depend on an organization's configuration and customization.
Searchable Fields: Business Brand: Searchable fields for Business Brand Salesforce Classic include various standard and custom fields that are relevant to the business brand. These fields can depend on an organization's configuration and customization.
Searchable Fields: Campaign: Searchable fields for Campaign Salesforce Classic include various standard and custom fields that are relevant to the campaign. These fields can depend on an organization's configuration and customization.
Searchable Fields: Care Metric Target: Searchable fields for Care Metric Target Salesforce Classic include various standard and custom fields that are relevant to the care metric target. These fields can depend on an organization's configuration and customization.
Searchable Fields: Care Pgm Prov Healthcare Provider: Searchable fields for Care Pgm Prov Healthcare Provider Salesforce Classic include various standard and custom fields that are relevant to the care pgm prov healthcare provider. These fields can depend on an organization's configuration and customization.
Searchable Fields: Care Program Enrollment Card: Searchable fields for Care Program Enrollment Card Salesforce Classic include various standard fields that are relevant to the care program enrollment card. These fields can depend on an organization's configuration and customization.
Searchable Fields: Care Program Provider: Searchable fields for Care Program Provider Salesforce Classic include various standard fields that are relevant to the care program provider. These fields can depend on an organization's configuration and customization.
Searchable Fields: Care Provider Adverse Action: Searchable fields for Care Provider Adverse Action Salesforce Classic include various standard fields that are relevant to the care provider adverse action. These fields can depend on an organization's configuration and customization.
Searchable Fields: Care Registered Device: Searchable fields for Care Registered Device Salesforce Classic include various standard and custom fields that are relevant to the care registered device. These fields can depend on an organization's configuration and customization.
Searchable Fields: Case: Searchable fields for Case Salesforce Classic include various standard and custom fields that are relevant to the case. These fields can depend on an organization's configuration and customization.
Searchable Fields by Object in Salesforce Classic: Your search results for a particular object depend on two factors: the type of search and the searchable fields for that object. Salesforce Classic has several search types—sidebar, advanced, global, and lookup.— Each type searches a unique set of fields for each object. Learn which search types and searchable fields are available for a given object.
Searchable Fields: Channel Program: Searchable fields for Channel Program Salesforce Classic include various standard and custom fields that are relevant to the channel program. These fields can depend on an organization's configuration and customization.
Searchable Fields: Channel Program Level: Searchable fields for Channel Program Level Salesforce Classic include various standard and custom fields that are relevant to the channel program level. These fields can depend on an organization's configuration and customization.
Searchable Fields: Chatter Feed: Searchable fields for Chatter Feed Salesforce Classic include various standard fields that are relevant to the chatter feed. These fields can depend on an organization's configuration and customization.
Searchable Fields: Chatter Group: Searchable fields for Chatter Group Salesforce Classic include various standard fields that are relevant to the chatter group. These fields can depend on an organization's configuration and customization.
Searchable Fields: Topic: Searchable fields for Topic Salesforce Classic include various standard fields that are relevant to the topic. These fields can depend on an organization's configuration and customization.
Searchable Fields: Coaching: Searchable fields for Coaching Salesforce Classic include various standard and custom fields that are relevant to coaching. These fields can depend on an organization's configuration and customization.
Searchable Fields: Code Set: Searchable fields for Code Set Salesforce Classic include various standard and custom fields that are relevant to the code set. These fields can depend on an organization's configuration and customization.
Searchable Fields: Comm Subscription: Searchable fields for Comm Subscription Salesforce Classic include various standard and custom fields that are relevant to the comm subscription. These fields can depend on an organization's configuration and customization.
Searchable Fields: Comm Subscription Channel Type: Searchable fields for Comm Subscription Channel Type Salesforce Classic include various standard and custom fields that are relevant to the comm subscription channel type. These fields can depend on an organization's configuration and customization.
Searchable Fields: Comm Subscription Consent: Searchable fields for Comm Subscription Consent Salesforce Classic include various standard and custom fields that are relevant to the comm subscription consent. These fields can depend on an organization's configuration and customization.
Searchable Fields: Comm Subscription Timing: Searchable fields for Comm Subscription Timing Salesforce Classic include various standard and custom fields that are relevant to the comm subscription timing. These fields can depend on an organization's configuration and customization.
Searchable Fields: Community: Searchable fields for Community Salesforce Classic include a standard field that is relevant to the community. This field can depend on an organization's configuration and customization.
Searchable Fields: Contact: Searchable fields for Contact Salesforce Classic include various standard and custom fields that are relevant to the contact. These fields can depend on an organization's configuration and customization.
Searchable Fields: Contact Point Type Consent: Searchable fields for Contact Point Type Consent Salesforce Classic include various standard and custom fields that are relevant to the contact point type consent. These fields can depend on an organization's configuration and customization.
Searchable Fields: Salesforce CRM Content: Searchable fields for Salesforce CRM Content Salesforce Classic include various standard and custom fields that are relevant to the Salesforce CRM content. These fields can depend on an organization's configuration and customization.
Searchable Fields: Contract: Searchable fields for Contract Salesforce Classic include various standard and custom fields that are relevant to the contract. These fields can depend on an organization's configuration and customization.
Searchable Fields: Contract Line Item: Searchable fields for Contract Line Item Salesforce Classic include various standard fields that are relevant to the contract line item. These fields can depend on an organization's configuration and customization.
Searchable Fields: Credit Memo Inv Application: Searchable fields for Credit Memo Inv Application Salesforce Classic include various standard and custom fields that are relevant to the credit memo inv application. These fields can depend on an organization's configuration and customization.
Searchable Fields: Custom Object: A custom object that’s associated with a custom tab is searchable by default, even if users don’t add the tab for display. By default, search is disabled for new custom objects. Admins enable Allow Search when setting up new custom objects.
Searchable Fields: Customer: Searchable fields for Customer Salesforce Classic include various standard and custom fields that are relevant to the customer. These fields can depend on an organization's configuration and customization.
Searchable Fields: Data Use Legal Basis: Searchable fields for Data Use Legal Basis Salesforce Classic include various standard and custom fields that are relevant to the data use legal basis. These fields can depend on an organization's configuration and customization.
Searchable Fields: Data Use Purpose: Searchable fields for Data Use Purpose Salesforce Classic include various standard and custom fields that are relevant to the data use purpose. These fields can depend on an organization's configuration and customization.
Searchable Fields: D&B Company: Searchable fields for D&B Company in Salesforce Classic and Data.com Prospector licenses. This includes sidebar, advanced, and global search, as well as fields such as D-U-N-S number, URL, and telephone number. Access to D&B Company records requires Data.com Prospector or Data.com Clean.
Searchable Fields: Document: Searchable fields for Document Salesforce Classic include various standard and custom fields that are relevant to the document. These fields can depend on an organization's configuration and customization.
Searchable Fields: Email Message: Searchable fields for Email Message Salesforce Classic include various standard and custom fields that are relevant to the email message. These fields can depend on an organization's configuration and customization.
Searchable Fields: Engagement Channel Type: Searchable fields for Engagement Channel Type Salesforce Classic include various standard and custom fields that are relevant to the engagement channel type. These fields can depend on an organization's configuration and customization.
Searchable Fields: Entitlement: Searchable fields for Entitlement Salesforce Classic include various standard and custom fields that are relevant to the entitlement. These fields can depend on an organization's configuration and customization.
Searchable Fields: Expense: Searchable fields for Expense Salesforce Classic include various standard and custom fields that are relevant to the expense. These fields can depend on an organization's configuration and customization.
Searchable Fields: Expense Report: Searchable fields for Expense Report Salesforce Classic include various standard and custom fields that are relevant to the expense report. These fields can depend on an organization's configuration and customization.
Searchable Fields: Expense Report Entry: Searchable fields for Expense Report Entry Salesforce Classic include various standard and custom fields that are relevant to the expense report entry. These fields can depend on an organization's configuration and customization.
Searchable Fields: External Object: Searchable fields for External Object Salesforce Classic include a custom field that is relevant to the external object. This field can depend on an organization's configuration and customization.
Searchable Fields: Feed Comment: Searchable fields for Feed Comment Salesforce Classic include various custom fields that are relevant to the feed comment. These fields can depend on an organization's configuration and customization.
Searchable Fields: Feed Post: Searchable fields for Feed Post Salesforce Classic include various custom fields that are relevant to the feed post. These fields can depend on an organization's configuration and customization.
Searchable Fields: File: Searchable fields for File Salesforce Classic include various standard and custom fields that are relevant to the file. These fields can depend on an organization's configuration and customization.
Searchable Fields: Goal: Searchable fields for Goal Salesforce Classic include various standard and custom fields that are relevant to the goal. These fields can depend on an organization's configuration and customization.
Searchable Fields: Gtwy Prov Payment Method Type: Searchable fields for Gtwy Prov Payment Method Type Salesforce Classic include various standard and custom fields that are relevant to the gtwy prov payment method type. These fields can depend on an organization's configuration and customization.
Searchable Fields: Idea: Searchable fields for Idea Salesforce Classic include various standard fields that are relevant to the idea. These fields can depend on an organization's configuration and customization.
Searchable Fields: Individual: Searchable fields for Individual Salesforce Classic include various standard and custom fields that are relevant to the individual. These fields can depend on an organization's configuration and customization.
Searchable Fields: Lead: Searchable fields for Lead Salesforce Classic include various standard and custom fields that are relevant to the lead. These fields can depend on an organization's configuration and customization.
Searchable Fields: List Email: Searchable fields for List Email Salesforce Classic include various standard and custom fields that are relevant to the list email. These fields can depend on an organization's configuration and customization.
Searchable Fields: Live Chat Transcript: Searchable fields for Live Chat Transcript Salesforce Classic include various standard fields that are relevant to the live chat transcript. These fields can depend on an organization's configuration and customization.
Searchable Fields: Location: Searchable fields for Location Salesforce Classic include various standard and custom fields that are relevant to the location object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Macro: Searchable fields for Macro Salesforce Classic include various standard fields that are relevant to the macro. These fields can depend on an organization's configuration and customization.
Searchable Fields: Maintenance Plan: Searchable fields for Maintenance Plan Salesforce Classic include various standard and custom fields that are relevant to the maintenance plan object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Metric: Searchable fields for Metric Salesforce Classic include various standard and custom fields that are relevant to the metric. These fields can depend on an organization's configuration and customization.
Searchable Fields: Note: Searchable fields for Note Salesforce Classic include various standard fields that are relevant to the note. These fields can depend on an organization's configuration and customization.
Searchable Fields: Operating Hours: Searchable fields for Operating Hours Salesforce Classic include various standard and custom fields that are relevant to the operating hours. These fields can depend on an organization's configuration and customization.
Searchable Fields: Opportunity: Searchable fields for Opportunity Salesforce Classic include various standard and custom fields that are relevant to the opportunity. These fields can depend on an organization's configuration and customization.
Searchable Fields: Order: Searchable fields for Order Salesforce Classic include various standard and custom fields that are relevant to the order. These fields can depend on an organization's configuration and customization.
Searchable Fields: Partner Fund Allocation: Searchable fields for Partner Fund Allocation Salesforce Classic include various standard and custom fields that are relevant to the partner fund allocation. These fields can depend on an organization's configuration and customization.
Searchable Fields: Partner Fund Claim: Searchable fields for Partner Fund Claim Salesforce Classic include various standard and custom fields that are relevant to the partner fund claim. These fields can depend on an organization's configuration and customization.
Searchable Fields: Partner Fund Request: Searchable fields for Partner Fund Request Salesforce Classic include various standard and custom fields that are relevant to the partner fund request. These fields can depend on an organization's configuration and customization.
Searchable Fields: Partner Marketing Budget: Searchable fields for Partner Marketing Budget Salesforce Classic include various standard and custom fields that are relevant to the partner marketing budget. These fields can depend on an organization's configuration and customization.
Searchable Fields: Party Consent: Searchable fields for Party Consent Salesforce Classic include various standard and custom fields that are relevant to the party consent. These fields can depend on an organization's configuration and customization.
Searchable Fields: People: Searchable fields for People Salesforce Classic include various standard and custom fields that are relevant to people. These fields can depend on an organization's configuration and customization.
Searchable Fields: Performance Cycle: Searchable fields for Performance Cycle Salesforce Classic include various standard and custom fields that are relevant to the performance cycle. These fields can depend on an organization's configuration and customization.
Searchable Fields: Person Account: Searchable fields for Person Account Salesforce Classic include various standard and custom fields that are relevant to the person account. These fields can depend on an organization's configuration and customization.
Searchable Fields: Price Book: Searchable fields for Price Book Salesforce Classic include various standard fields that are relevant to the price book. These fields can depend on an organization's configuration and customization.
Searchable Fields: Product: Searchable fields for Product Salesforce Classic include various standard and custom fields that are relevant to the product. These fields can depend on an organization's configuration and customization.
Searchable Fields: Product Consumed: Searchable fields for Product Consumed Salesforce Classic include various standard and custom fields that are relevant to the product consumed object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Product Item: Searchable fields for Product Item Salesforce Classic include various standard and custom fields that are relevant to the product item object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Product Request: Searchable fields for Product Request Salesforce Classic include various standard and custom fields that are relevant to the product request object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Product Request Line Item: Searchable fields for Product Request Line Item Salesforce Classic include various standard and custom fields that are relevant to the product request line item object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Product Transfer: Searchable fields for Product Transfer Salesforce Classic include various standard and custom fields that are relevant to the product transfer object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Product Warranty Term: Searchable fields for Product Warranty Term Salesforce Classic include various standard and custom fields that are relevant to the product warranty term. These fields can depend on an organization's configuration and customization.
Searchable Fields: Provider Search Sync Log: Searchable fields for Provider Search Sync Log Salesforce Classic include various standard and custom fields that are relevant to the provider search sync log. These fields can depend on an organization's configuration and customization.
Searchable Fields: Question: Searchable fields for Question Salesforce Classic include various standard fields that are relevant to the question. These fields can depend on an organization's configuration and customization.
Searchable Fields: Quick Text: Searchable fields for Quick Text Salesforce Classic include various standard and custom fields that are relevant to the quick text. These fields can depend on an organization's configuration and customization.
Searchable Fields: Quote: Searchable fields for Quote Salesforce Classic include various standard and custom fields that are relevant to the quote. These fields can depend on an organization's configuration and customization.
Searchable Fields: Report: Searchable fields for Report Salesforce Classic include various standard fields that are relevant to the report. These fields can depend on an organization's configuration and customization.
Searchable Fields: Resource Absence: Searchable fields for Resource Absence Salesforce Classic include various standard and custom fields that are relevant to the resource absence. These fields can depend on an organization's configuration and customization.
Searchable Fields: Return Order Line Item: Searchable fields for Return Order Line Salesforce Classic include various standard and custom fields that are relevant to the return order line. These fields can depend on an organization's configuration and customization.
Searchable Fields: Return Order: Searchable fields for Return Order Salesforce Classic include various standard and custom fields that are relevant to the return order. These fields can depend on an organization's configuration and customization.
Searchable Fields: Reward Fund: Searchable fields for Reward Fund Salesforce Classic include various standard and custom fields that are relevant to the reward fund. These fields can depend on an organization's configuration and customization.
Searchable Fields: Reward Fund Type: Searchable fields for Reward Fund Type Salesforce Classic include various standard and custom fields that are relevant to the reward fund type. These fields can depend on an organization's configuration and customization.
Searchable Fields: Seller: Searchable fields for Seller Salesforce Classic include various standard and custom fields that are relevant to the seller. These fields can depend on an organization's configuration and customization.
Searchable Fields: Service Appointment: Searchable fields for Service Appointment Salesforce Classic include various standard and custom fields that are relevant to the service appointment. These fields can depend on an organization's configuration and customization.
Searchable Fields: Service Contract: Searchable fields for Service Contract Salesforce Classic include various standard fields that are relevant to the service contract. These fields can depend on an organization's configuration and customization.
Searchable Fields: Service Crew: Searchable fields for Service Crew Salesforce Classic include various standard and custom fields that are relevant to the service crew. These fields can depend on an organization's configuration and customization.
Searchable Fields: Service Crew Member: Searchable fields for Service Crew Member Salesforce Classic include various standard and custom fields that are relevant to the service crew member. These fields can depend on an organization's configuration and customization.
Searchable Fields: Service Resource: Searchable fields for Service Resource Salesforce Classic include various standard and custom fields that are relevant to the service resource. These fields can depend on an organization's configuration and customization.
Searchable Fields: Service Resource Capacity: Searchable fields for Service Resource Capacity Salesforce Classic include various standard and custom fields that are relevant to the asset object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Service Resource Skill: Searchable fields for Service Resource Skill Salesforce Classic include various standard and custom fields that are relevant to the service resource skill. These fields can depend on an organization's configuration and customization.
Searchable Fields: Service Territory: Searchable fields for Service Territory Salesforce Classic include various standard and custom fields that are relevant to the service territory. These fields can depend on an organization's configuration and customization.
Searchable Fields: Service Territory Member: Searchable fields for Service Territory Member Salesforce Classic include various standard and custom fields that are relevant to the service territory member. These fields can depend on an organization's configuration and customization.
Searchable Setup Objects: Use global search in Setup to find specific setup records, such as the Lead Source picklist or the Sales Rep profile. Global search differs from Quick Find, which finds pages within the Setup menu, such as Account Settings or Profiles.
Searchable Fields: Shipment: Searchable fields for Shipment Salesforce Classic include various standard and custom fields that are relevant to the shipment object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Skill: Searchable fields for Skill Salesforce Classic include various standard and custom fields that are relevant to the skill. These fields can depend on an organization's configuration and customization.
Searchable Fields: Skill Requirement: Searchable fields for Skill Requirement Salesforce Classic include various standard and custom fields that are relevant to the skill requirement object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Social Persona: Searchable fields for Social Persona Salesforce Classic include a field that is relevant to the social persona. This field can depend on an organization's configuration and customization.
Searchable Fields: Social Post: Searchable fields for Social Post Salesforce Classic include various standard fields that are relevant to the social post. These fields can depend on an organization's configuration and customization.
Searchable Fields: Solution: Searchable fields for Solution Salesforce Classic include various standard and custom fields that are relevant to the solution. These fields can depend on an organization's configuration and customization.
Searchable Fields: Survey: Searchable fields for Survey Salesforce Classic include various standard and custom fields that are relevant to the survey. These fields can depend on an organization's configuration and customization.
Searchable Fields: Time Sheet: Searchable fields for Time Sheet Salesforce Classic include various standard and custom fields that are relevant to the time sheet object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Time Sheet Entry: Searchable fields for Time Sheet Entry Salesforce Classic include various standard and custom fields that are relevant to the time sheet entry object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Unit Of Measure: Searchable fields for Unit Of Measure Salesforce Classic include various standard and custom fields that are relevant to the unit of measure. These fields can depend on an organization's configuration and customization.
Searchable Fields: User: Searchable fields for User Salesforce Classic include various standard and custom fields that are relevant to the user. These fields can depend on an organization's configuration and customization.
Searchable Fields: WDC Feedback: Searchable fields for WDC Feedback Salesforce Classic include various standard fields that are relevant to the WDC feedback. These fields can depend on an organization's configuration and customization.
Searchable Fields: Work Order: Searchable fields for Work Order Salesforce Classic include various standard and custom fields that are relevant to the work order. These fields can depend on an organization's configuration and customization.
Searchable Fields: Work Order Line Item: Searchable fields for Work Order Line Item Salesforce Classic include various standard and custom fields that are relevant to the work order line item. These fields can depend on an organization's configuration and customization.
Searchable Fields: Work Type: Searchable fields for Work Type Salesforce Classic include various standard and custom fields that are relevant to the work type object. These fields can depend on an organization's configuration and customization.
Searchable Fields: Work Type Group: Searchable fields for Work Type Group Salesforce Classic include various standard and custom fields that are relevant to the work type group. These fields can depend on an organization's configuration and customization.
Customize Search with Search Manager: Customize the search experience for different channels and user needs. Provide relevant results and manage your configurations.
Field-Level Security Applied in Search: FAQ: Before the Summer 24 release, Salesforce search results sometimes returned unexpected outcomes under certain circumstances due to a limitation in how Search applies Field Level Security. This limitation arises because Search applies Field-Level Security to all standard fields and the first 100 searchable custom fields, but not to custom fields beyond the 100th. Therefore, searches can use these unprotected fields to return results.
Einstein Search Terms: Learn the definitions of terms associated with Einstein Search and how they apply to the search experience.
Objects and Languages Affected: Lemmatization, synonym groups, spelling correction, and nicknames all affect different objects. Review each term, and see which objects and languages it searches on.
How Einstein Search Works: From the moment you create a record in Salesforce, the search engine works to make your information easy to find. Get a high-level look at how Einstein Search works.
Search Index: The search index makes it quicker and more efficient for the search engine to retrieve relevant information for a user’s query. Without a search index, the search engine scans all content in every document or record in your organization each time a user performs a search, which is time-consuming and resource-intensive. Explore how the search index can improve your search engine.
Make Search Faster: In Salesforce Classic, disabling search for custom and external objects, and scheduling bulk uploads during off-peak hours helps speed up search.
Manage the Search Index: View default searchable fields. You can add and remove fields in the search index. For example, if you never want to include certain custom fields in search, remove those fields. Removed fields don’t appear in search results and aren’t used to match results. You can’t remove standard fields.
Get Instant Results in Salesforce Classic: As you type in the global search bar, a list of suggested records appears. Use these instant results for quick record access. If you don’t see the record that you want in the list, perform a full search.
Search Interactions DMO: The Search Interactions DMO is a Data Cloud data model object (DMO) for a specific interaction on a search result, such as clicks, hovers, and attachments.
Additional Layout Configurations: Customize your users search layouts further by including more filtering and sorting options, and modifying Action buttons and Related List Quick Links.
Search Layout Guidelines: Before customizing your layouts, keep in mind the fields that are included in a search layout.
Customize Search for Service Agents: Improve search efficiency and help agents find the relevant information in your knowledge base faster. For example, optimize the search for agents in a specific product area and apply it to the Knowledge component for that product area. Choose which fields are available to search, and filter out results they don’t need.
Lookup Search in Salesforce Classic: Lookup fields allow you to associate two records together in a relationship. For example, a contact record includes an account lookup field that associates the contact with its account.
Secondary Field Usage and Limitations: The secondary field provides context to help you determine which record is the one you’re looking for. From the global search bar, the secondary field is beneath the record name and next to the object type. For lookups, it’s below the record name. Advanced lookup matches results against all searchable fields in a record, including the secondary field, while lookup suggestions don’t.
Lookup Searches: With lookup fields, you associate two records to create a relationship. For example, a contact record includes an account lookup field that associates the contact with its account.
Standard and Enhanced Lookups in Salesforce Classic: Salesforce objects often include lookup fields that allow you to associate two records together in a relationship. For example, a contact record includes an Account lookup field that associates the contact with its account. Lookup fields appear with the search icon on record edit pages. Clicking the icon opens a lookup search dialog that allows you to search for the record that you want to associate with the record you’re editing. There are two main types of lookups: standard and enhanced.
Enhanced Lookups in Salesforce Classic: Enhanced lookups offer more search capabilities than standard lookups. In addition, enhanced lookups allow search result sorting and filtering.
Standard Lookups in Salesforce Classic: Standard lookups search through a limited set of searchable fields per object and users can’t refine search results.
Lookup Search Cheat Sheet: Quickly compare features between Lightning Experience and the Salesforce Mobile app.
How to Make Search Work Faster: Disabling search for custom and external objects and scheduling bulk uploads during off-peak hours helps speed up search.
Customize Global Search for Your Users: Optimize global search for each user profile’s needs and help them find relevant results quickly. For example, your sales agents search for customer contacts and opportunities, while marketing agents primarily search for campaigns and leads. Customize search differently for these two user profiles.
Considerations for Rules: Keep a few considerations in mind when defining rules in a Search Manager configuration.
Customize Search for Your Users: In your Salesforce org, users search for different purposes in each search channel. Create separate configurations for users in each search channel. Apply rules to optimize search and select objects to always search.
Customize Explicit Search Filters: In Einstein Search, you can define explicit search filters that narrow the results based on certain criteria. To further tailor the search experience for specific users, you can customize a default explicit search filter based on a profile for an object. For example, you can customize the search filters that are available for the Contacts object specifically for the Contract Manager profile.
Delete a Search Configuration: Delete unused configurations. You can’t delete LWR Experience site configurations that are currently used in a site.
Filter Search with Rules: Filter out irrelevant results before your users see them by adding rules to a search configuration. Rules are hidden filters that define the scope of the search results to show.
Manage Search Configurations: Clear up clutter by editing, deleting, and migrating Search Manager configurations.
Manage Searchable Objects and Fields: Optimize search results and make search more efficient by excluding objects and fields that aren’t important to your users. You can manage searchable objects for each profile in global search and for Experience Cloud configurations. You can manage searchable custom fields for all configurations.
Optimize Search Results: Ensure that users see relevant results. Manage searchable objects and fields, filter out irrelevant results with rules, and select which objects to always search.
Target a Search Configuration for a Channel: Create a configuration to target specific user needs in global search, users in Experience Cloud sites, or service agents in the Knowledge console. Select searchable objects and fields, and optimize search.
Work Faster and Smarter with Einstein Search: Einstein Search is personal, natural, and actionable. Einstein Search gives you the power to find information faster so that you can work more effectively. You can act quickly with search results that are more actionable and tailored to how you work in Salesforce.
Field-Level Security for Custom Fields in Search: Field-level security specifies whether users in your Salesforce org can view and edit each field in an object. You can configure field-level security settings separately from Einstein Search. Users can search a protected field only if they have the appropriate field permissions. However, search can only protect up to 100 searchable custom fields per object with field-level security, even if you set field-level security for more than 100 custom fields. The search engine matches results based on information in unprotected fields. You can select which custom fields are protected.
Search Queries DMO: The Search Queries DMO is a Data Cloud data model object (DMO) for a specific search query, such as the date of the query and the number of results the query found.
Search Query Limits: Limits on search queries protect against service disruptions and help ensure optimal performance for normal usage.
How Search Results are Ordered: The search engine sorts results most relevant to the user based on search term frequency, order, uniqueness, record activity, and access permissions. Search result relevance can vary for each user. Search returns only the records that you have access to view.
Search Results DMO: The Search Results DMO is a Data Cloud data model object (DMO) for a specific search result for a query.
Configure Einstein Search: Grant your users access to Einstein Search. From there, you can unlock new capabilities that can simplify your search process and customize your users profiles.
Right-to-Left (RTL) Language Limitations: For right-to-left language users, search supports Hebrew and Arabic as end-user languages and Urdu as a platform language. For these languages, Salesforce provides limited support for global search and lookup search features in Lightning Experience.
Build Your Synonym Groups: A search for one term in a synonym group returns results for all terms in the group. For example, a search for thumb drive returns results for flash drive and memory stick when these three terms are in a synonym group.
How Search Breaks Up Information: To efficiently store and find information, the search engine breaks up record content into smaller pieces, called tokens. The tokens are stored in the search index. See how common types of information, including emails, websites, and phone numbers are tokenized.
Search for Records: Your search for records begins with the global search bar at the top of each page. From the moment you click the search bar, search starts delivering results.
Configure Search Layouts in Salesforce Classic: Search layouts let you determine what users see when search returns results. You can create search layouts for standard and custom objects, ensuring the layout shows users what’s most relevant to them. Layouts are applied on a per-object basis. The modifications you make for an object’s layout applies to all users.
Refine Search With Wildcards and Operators: Get more control over your search results by using wildcards and operators in your searches. Using wildcards and operators in your searches can ensure you’re seeing all applicable records in your search results. Learn how to use them and their limitations.
Considerations for Agentforce Service Agent: To use Agentforce Service Agent, consider supported functionality, usage, limitations and allowances, limits, and other issues.
Use Context Variables in Messaging Conversations: Personalize agent conversations over messaging channels with context variables. Context variables let your agent use Messaging Session record fields as action inputs.
Verify Customers with Standard Subagents in Agentforce Builder: To verify the identity of an unverified user in an agent session, configure your agent to use the Customer Verification subagent or the Service Customer Verification subagent and limit access to subagents and actions that you specify. You can require different levels of verification for different subagents and actions, based on your business’s security requirements.
Connect an Agent to Enhanced Chat v2 in Agentforce Builder: To connect an agent to Enhanced Chat v2 in the new Agentforce Builder, set up the Enhanced Chat v2 connection. Agentforce Builder creates the channel and Enhanced Chat v2 deployment for you. Set up Agentforce Voice for Enhanced Chat v2 so that users can switch between speech and text when they chat with an Agentforce Service agent.
Connect a Service Agent to Enhanced Chat v2 in the Legacy Builder: To connect an agent to Enhanced Chat v2 in the legacy Agentforce Builder, set up the Enhanced Chat v2 connection. Route messages to a Service agent directly or with an inbound Omni-Channel flow.
Update Your Routing Logic for Agentforce Service Agent on Email: Nearly all of the puzzle pieces are in place, but Salesforce still needs to know when the agent should use your email template to email customers. We recommend creating an Omni-Channel flow that controls which cases are routed to your Agentforce Service agent versus a service rep.
Test Agentforce Service Agent on Email: If you’re using an Omni-Channel flow for routing, test your flow with a new unused routing address. Then, switch to an active routing address when you’re ready.
Use Custom Variables for Agentforce Service Email: Agentforce Service Agent now passes inbound email context, such as the Case ID, into custom variables instead of bundling it with the user message. You can reference these variables in your topics and action inputs for more reliable automation.
Transfer Conversations from an Agent with an Omni-Channel Flow: When an agent encounters conversations that it can’t resolve, it uses the Escalation subagent to escalate the conversation. In the new Agentforce Builder, you can also run the Escalation subagent using the escalate utility function in Agent Script. Escalation and transferring works differently depending on the agent type and channel. You can customize the experience that customers receive when transferring isn’t possible.
Use Additional Languages in Enhanced Chat Conversations: Reach more customers and employees with a multi-language agent. After you select additional languages in your agent’s language settings and set up an Enhanced Chat channel, finish setting up additional languages for your agent.
Configure Service Agent Managers: When Service agents are enabled for your Salesforce org, Salesforce admins can create and edit service agents. To define other users as Service agent managers so that they can also create and edit service agents, assign those users the Manage Agentforce Service Agents permission set.
Show Progress Indicators in Bring Your Own Channel: Progress indicators tell the messaging user what actions an Agentforce Service agent is taking behind the scenes of a Bring Your Own Channel messaging session. Standard actions come with pre-written progress indicators, such as "Doing research" or "Analyzing". To customize the progress indicator for an action, edit its loading text.
Customize Loading Text for Enhanced Chat: Loading text tells the user what an agent is doing while an action runs in the background. To show your users loading text in Enhanced Chat v1, set up typing indicators. Users see loading text by default on Enhanced Chat v2 channels. Optionally, customize the loading text color and label on the Embedded Service Deployments Settings Setup page.
Build a Help Agent in Minutes with Quick Service Agent Configuration: In a few quick steps, build AI-powered service channels grounded in your knowledge base by using Quick Service Agent Configuration. This process can build your help portal, Voice channel, and web chat channel. It also builds a Help Agent to resolve customer issues in those channels. Use these channels to provide 24/7 support and deflect cases.
Maintain Trust with Agentforce Actions in the Legacy Builder: Agents connected to employee channels, such as Lightning Experience, the Salesforce mobile app, and Slack, limit access to subagents and actions by default, based on the end user's context. Agentforce Service agents connected to customer channels, such as enhanced Messaging channels and Enhanced Chat, require additional configuration to ensure secure access. As you configure your agent, it's important to think through the security and identification requirements for your use case and only grant your agents the actions and access required to complete the tasks you want them to be able to complete autonomously on behalf of your customers. For sensitive actions, we recommend building customer verification directly into the action’s flow.
Example Implementation of a Secure Agentforce Service Agent: The details of implementing secure Agentforce Service agents vary depending on where the agent is deployed and the surrounding user authentication policies. To get a feel for the underlying patterns, let’s look at an example.
Add User Identification to Agentforce Actions: When configuring your agent that connects to customer channels so that it can take private actions securely, have it confirm the identity of the person it’s interacting with according to your company’s standards.
Create an Agent from an Agentforce Service Agent Template: Agentforce Service agents intelligently support your customers by connecting to messaging and other channels and escalating to reps when necessary. Use the default Agentforce Service Agent template to create an agent designed to resolve common support cases and requests. Or select from more specialized templates to fit your use case.
Create a Voice-Enabled Service Agent: To create a voice-enabled agent, create a new agent, add a voice-enabled connection, such as Telephony and configure voice mode settings.
Verify Customers with Standard Subagents in the Legacy Builder: Before an Agentforce Service agent takes a private action on a user’s behalf, Salesforce recommends verifying the identity of the person the agent is interacting with. To verify the identity of an unverified user in an agent session, configure your agent to use the Customer Verification subagent or the Service Customer Verification subagent and limit access to subagents and actions that you specify. You can require different levels of verification for different subagents and actions, based on your business’s security requirements.
Set Tool Restrictions for the Salesforce MCP Server: Tool restrictions control what Claude is allowed to do in your org through the Salesforce MCP connection. Most of these settings let skills discover, describe, and read your org's data, and must be set to Always Allowed for skills to work at all.
Setup | Get and Explain Object Permissions of User (Beta): Get the specified user’s object permissions for a given object. This action returns a list of the user’s assigned object permissions, which can include Create, Read, Edit, Delete, View All, and Modify All. It also can get the profile, permission sets, or permission set groups that grant the user this access.
Setup | Get and Explain User Permissions of User (Beta): Get whether a specified user is assigned the required user permissions for a given task or feature. This action can also return the profile, permission sets, or permission set groups that grant the user this permission, if applicable.
Setup | Identify Field By Name: This action interprets the user’s input to determine if the field exists within an object.
Setup | Identify Object By Name (Object Management): Admins can use this action under Object Management to interpret user input and determine which object the user is referring to. It returns a list of all matching objects available under Object Manager.
Setup | Migrate Connected App: Identifies the configurations of a connected app and creates an identical external client app to replace it. After connected app migration, the connected app remains in the App Manager list but is read only.
Agentforce Platform | Answer Questions with Salesforce Documentation (Beta): Answers a question from a user based on information from relevant Salesforce documentation, including links to source articles. This action is the fallback solution if a user makes a request in Setup to the agent and there’s no action available to complete that request. This action can also be added to other agents, such as the Employee agent in Lightning Experience.
Setup | Summarize OAuth in Connected App: Generates a summary of the OAuth configurations for a connected app. The summary output includes general information about the connected app as well as configurations and descriptions of all enabled OAuth Settings and Policies fields. This action is related to the Migrate Connected App feature.
Add Context to a Skill: Add reference data so the skill is grounded in your org at run time.
Enable Skill Builder: Assign the AI Skill User or AI Skill Manager permission set to access Skill Builder in the App Launcher, Agentforce Studio, and Agentforce Coworker.
Skill Builder Examples: Copy a sample skill into Skill Builder, and then adapt the name, description, and instructions for your org. The Daily Briefing sample returns a salesperson’s meetings, pipeline, cases, and Slack highlights.
Give Users Access to Skills: Assign AI Skill User or AI Skill Manager, and then decide who can use each skill.
How Users Work with a Skill: After you publish a skill, users run it from Use in chat, from a slash command, or by describing a task.
Skills: Create, test, publish, and share skills in Skill Builder so an AI agent can complete a repeatable task.
Key Concepts: Learn the key terms used in Skills and Skill Builder.
Manage Skills: Share skills and choose which published skills appear in My Skills.
Troubleshoot Skills: Resolve common issues when a skill doesn’t appear or doesn’t run as expected.
Skill Versions: A skill version is a snapshot of a skill’s instructions and configuration. The published version is the one that runs.
Write Skill Instructions: Write Markdown instructions that tell the AI model what to do, in what order, and how to use agent actions and context.
Find Records Using Salesforce Classic Tags: From the Tags page, you can view, browse, search tags, find records with tags, and modify or remove tags. You can also view the number and use of personal tags.
Configure Call Escalation When Using PSTN: If the agent call is escalated, the agent is disconnected and the telephony system resumes the call. If supported, configure the telephony system to handle the escalated call, including transferring the call to the appropriate rep.
Configure Call Escalation When Using Dynamic Routing: If the agent call is escalated to a rep, the agent is disconnected and the telephony system resumes the call. If supported, configure the telephony system to handle the escalated call, including resuming the call when it's escalated and transferring the call to the appropriate rep.
Configure Call Escalation When Using Real-Time Streaming: If the agent call is escalated to a rep, the agent is disconnected and the telephony system resumes the call. If supported, configure the telephony system to handle the escalated call, including resuming the call when it’s escalated and transferring the call to the appropriate rep.
Configure Call Escalation When Using SIP: If the agent call is escalated to a rep, the agent is disconnected and the telephony system resumes the call. If supported, configure the telephony system to handle the escalated call, including resuming the call when it’s escalated and transferring the call to the appropriate rep. To illustrate how this works with SIP, this topic describes how to accomplish this task using Salesforce Voice with Partner Telephony from Genesys. Specifically, configure the Genesys inbound flow to resume the call where it left off, which is after the original call is transferred to the agent, and route it to a service rep.
Configure Inbound Call Transfers Using PSTN: To illustrate how to configure inbound call transfers that happen via PSTN, this page describes how to accomplish these tasks using Amazon Connect.
Configure Inbound Call Transfers When Using SIP: To illustrate how to configure inbound call transfers that happen via SIP, this topic describes how to accomplish these tasks with Genesys.
Configure SIP Trunking in Your Partner Telephony System: Before you can configure your telephony system to transfer and escalate calls via SIP, connect your telephony system to Salesforce. In addition to configuring logic to manage traffic and routing, connecting the telephony system to Salesforce also involves creating an encrypted SIP trunk to support user-to-user information exchange and SIP invites.
Procure a Phone Number to Transfer Calls Using PSTN: To transfer calls to the Agentforce agent using PSTN, first procure a phone number. Later, you associate this phone number with the channel that’s used to transfer calls to the agent.
Configure Telephony Vendor Settings When Using PSTN: If supported, configure your telephony system to create a VoiceCall record in Salesforce for every incoming call, and then transfer each call to the Agentforce agent using the procured phone number.
Configure Telephony Vendor Settings When Using Real-Time Streaming: If your telephony system supports it, configure it to create a VoiceCall record in Salesforce for each incoming call. Then transfer each call to the Agentforce agent by using the Real-Time Streaming phone number.
Configure Telephony Vendor Settings When Using SIP: If supported, configure your telephony system to create a VoiceCall record in Salesforce for every incoming call, and then transfer each call to the Agentforce agent using the procured SIP address.
CRM Analytics for Sales Cloud Einstein: Get access to more advanced analytics dashboards built on CRM Analytics. Salesforce admins can evaluate the effectiveness of Einstein features such as Einstein Lead Scoring. Sales leaders, managers, and reps can use Einstein Discovery for Reporting and Sales Analytics to get deeper insights into their sales data. From the Analytics tab or Analytics Studio, see the apps you have access to.
What Is Salesforce Platform: Salesforce Platform is the product that makes it possible for Claude and other AI tools to work directly with your Salesforce data, workflows, and permissions, without custom integration work. It's part of AIforce, Salesforce's underlying architecture for AI-to-data connections.
