AI agents that actually do the work

Agentforce deploys autonomous AI agents that triage, route, respond and act within your Salesforce environment. Not just suggest next steps. Actually execute them, within the guardrails you define.

Chatbots answer questions. Agents take action. Agentforce deploys AI agents across your Salesforce environment that can triage cases, draft responses, update records, and trigger workflows, autonomously, within the guardrails you define.

How Agentforce works

Incoming requests are processed by the Agentforce agent, which triages, routes, responds and acts across your Salesforce objects.

INCOMING REQUESTS Citizen query Portal submission Case update Status change trigger Document upload File attachment AI Agentforce agent Triage Route Respond Act Autonomous decision-making within guardrails SALESFORCE OBJECTS Cases Service Cloud Accounts Person and org Knowledge Article base Workflows Flows and automation Guardrails and governance Action boundaries | Audit logging | Escalation rules | Configurable permissions

What agents can do

Three use cases where Agentforce moves from suggestion to execution.

Case triage and routing

Incoming cases are automatically classified by type, urgency and department. The agent routes each case to the right team with full context attached, so nothing sits in a queue waiting for someone to read it first.

Citizen response drafting

The agent drafts responses using your knowledge base and case history. Human agents review and send, cutting response times significantly while maintaining quality and consistency.

AI-powered data extraction

Agents extract data from uploaded documents, validate it against existing records, and populate Salesforce objects without manual entry. Certificates, applications and supporting evidence are processed in seconds, not hours.

Why Agentforce is different

Actions, not answers

These agents don't just suggest next steps. They execute within defined boundaries: updating records, triggering workflows, routing cases and drafting responses autonomously.

Governed and auditable

Every action is logged, every decision is traceable, and every guardrail is configurable. You control what agents can and cannot do, with full visibility into what they actually did.

Built on your data

Agents work because the data underneath them is clean, connected and governed. MuleSoft integrates your systems, Informatica ensures data quality, and Agentforce acts on what it can trust.

Ready to put AI agents to work?

Let's talk about where Agentforce can make the biggest impact in your organisation.