Agentforce has the missing systems you need to make AI agents real for your business. You can get started today with prebuilt agents or customize your own. Agentforce agents have built-in integration with your business data and processes. They learn and adapt as they navigate to the desired outcome, all while seamlessly collaborating with employees inside the business apps they already use.
A Large Language Model alone is not enough. AI requires many systems to make it accurate, secure, and capable of taking action in your business. But thankfully, Agentforce provides those missing systems so you don't have to build them.
Simply type a natural-language description of the job you want your agent to do. Agent Builder will use an understanding of your business, powered by the metadata inside Salesforce, to auto-generate your first prototype. You can then easily monitor an Agent’s plan of action and test its responses right in Agent Builder.
Fine-tune your perfect agent with customization using natural-language instructions. These become the guardrails that direct your agent's behavior. This context helps your agent better understand the job it’s being asked to do.
Tap into the workflows and actions you’ve already built and optimized in Salesforce to expand the functionality of your agents. Equip your agents with those workflows to start working within your guidelines, from day 1.
Add further actions to your agents with AI capabilities. Use Prompt Builder templates to customize actions and automatically assign jobs to other agents.
Start answering customer questions and deflecting incoming support cases by using a prebuilt template to create your first Agentforce service agent. Your Agentforce service agent will be instantly aware of your customer records, case records, and the answers contained inside in your knowledge articles inside Salesforce. Which enables it to immediately start answering customer questions.
Nurture inbound leads and advance your pipeline with an SDR agent. Your agent will autonomously answer product questions, handle objections, and schedule meetings for your reps. The agent will engage leads in any language on their preferred channels, like SMS or WhatsApp by referencing CRM and external data. Your guardrails control when and how often it engages.
Practice pitching, handling objections, and negotiating with a sales coach agent that runs tailored role-plays. Get feedback on strengths, areas for improvement, and actionable steps based on CRM data, selling methodologies, and virtual meetings.
Offer AI-powered shopping assistance on commerce sites or messaging apps with your personal shopper agent. The agent uses CRM data to guide search and personalize product recommendations, boosting conversion rates. Customers can even add to their cart, check out, track orders, and receive updates without leaving the conversation.
Build and launch campaigns faster with a campaign agent that can generate briefs, audience segments, content, and customer journeys. The agent personalizes 1:1 engagements at scale using real-time data and continuously analyzes performance, proactively recommending improvements to optimize KPIs and drive better results.
With each new job, your AI agents will use a cycle of observations, evaluations, recommendations, and execution to carry out a job, step-by-step, learning and adapting to reach the right outcome.
Dramatically improve the accuracy of your agents with a semantic search to retrieve similar content, across all your data, to get the job done right. This advanced data retrieval technique (Retrieval Augmented Generation) is built-in to Agentforce, available to any AI agent.
Your agents will only be effective if they can tap into a much wider set of knowledge as they do work. The advantage of Agentforce agents is their built-in access to your CRM data, like customer information, opportunities, purchase and case history, campaigns, and more.
Agentforce agents also have built-in access to Data Cloud, so you can connect and harmonize all your external data without having to copy it. Plus, give your agents more context about what it all means with your metadata, the data about your data.
The who, what, and where behind an agent's job are critical to understanding it. Because Agentforce is built on the Salesforce Platform and part of your Customer 360, your agents get a full view of the who behind the job. Each time you create an agent, you’re also establishing a role for your agent, which will give that agent more context about what the job is. As for where, your agents can be embedded in your website, your mobile app, WhatsApp, in your employee apps, or inside your businesses processes.
All of your existing Salesforce security policies are automatically respected by your AI agents. You won’t need to recreate your existing security policies for AI. We’ve already negotiated zero-retention data agreements with our AI model partners, ensure that none of your data will ever be retained or used to enhance any AI results other than your own.
Your AI agents work inside the guardrails you establish to do jobs. They can reason intuitively to respond reliably to complex requests. More importantly, they will refuse to be lured outside those guardrails — or respond to any requests that might be inappropriate.
Agentforce automatically detects harm and toxicity when a request is inappropriate and will refuse to engage.
The Trust Layer identifies and masks sensitive information in your prompts before sending them to the model, protecting sensitive data like PII from being exposed to external LLMs. Ensure the safety and accuracy of your agent by tracking the functionality of its data masking and toxicity detection. You can delete your AI audit and feedback data at any time.
Take a closer look at how agent building works in our library.
Work with Professional Services experts to quickly build agents and see value.
Tell us about your business needs, and we’ll help you find answers.
Agentforce empowers businesses with autonomous AI agents that operate independently, adapt to real-time data, and seamlessly collaborate with human teams. Built on a low-code/no-code platform, Agentforce is adaptable to various industries and uses advanced AI capabilities to deliver accurate, efficient, and creative solutions.
Agentforce utilizes Salesforce's robust security infrastructure, including the Einstein Trust Layer. This ensures data privacy and limits access to authorized personnel. Advanced security measures, like data encryption and strict access controls, further safeguard sensitive information.
Salesforce also conducts regular security audits and adheres to industry standards to maintain a high level of security for Agentforce and its users.
No, you don't need to know how to code to use Agentforce. It's designed with a low-code/no-code approach, making it accessible to users with varying technical backgrounds. With its user-friendly interface, you can build and customize AI agents without writing complex code. That gives you the power to quickly deploy agents and use AI to automate tasks and improve efficiency.
There are so many ways you can use Agentforce to streamline operations, boost productivity, and improve customer experiences. Just some include:
Agentforce Service Agent and Agentforce are now generally available. Agentforce Service Agent pricing starts at $2 USD per conversation; standard volume discounts apply.
While copilots and chatbots are valuable tools, Agentforce offers a more advanced level of AI. It empowers agents with greater autonomy, enabling them to handle a wider range of tasks independently. With advanced AI capabilities, Agentforce agents can provide more accurate, efficient, and creative solutions, ultimately delivering greater value to businesses.
Agentforce is deeply integrated with the Salesforce Platform, allowing AI agents to seamlessly access and interact with data across various Salesforce applications. This integration enables agents to use customer data, trigger workflows, and update records, making them a powerful tool within the Salesforce ecosystem.
