Salesforce Unveils Expanded Headless 360 Platform for Secure AI Agent Interactions

Salesforce is expanding its Headless 360 platform to make it easier for AI agents to securely interact with enterprise software. The company's goal is to provide a secure foundation for the Agentic Enterprise, allowing authorized AI agents to discover, understand, and take action across the business without custom integrations or duplicated business logic.

The new Model Context Protocol (MCP) servers are built on open standards and enable agents running in various AI platforms to dynamically discover and invoke Salesforce capabilities in real-time. The Headless 360 MCP Server exposes trusted business capabilities, rather than APIs, and allows agents to understand the relationships, permissions, workflows, validation rules, and governance that define how the business operates.

The expansion of Headless 360 includes Data 360 capabilities, Slack integrations, developer tools, reusable Skills, and headless experiences across every major cloud. This milestone builds on Salesforce's long history of openness, making trusted enterprise capabilities available wherever AI agents work. The company is already seeing the impact of this innovation, with Engine launching its AI support agent, EVA, in just 12 days, which resolves half of all customer chat interactions without a human on the line.