Salesforce expands Headless 360 to support AI agents across enterprise applications
Salesforce has announced an expansion of its Headless 360 platform, adding capabilities intended to allow authorised AI agents to discover and use Salesforce business functions across enterprise applications.
The expansion includes new Model Context Protocol (MCP) servers, Data 360 capabilities, Slack integrations, reusable Skills and developer tools. The company said the approach is aimed at allowing organisations to reuse existing Salesforce business logic, permissions, workflows and governance rather than building separate integrations for individual AI applications.
At the centre of the announcement is the Headless 360 MCP Server, which uses the open MCP standard to allow AI agents to discover and invoke Salesforce capabilities. According to Salesforce, the server can be used with AI platforms including Agentforce, Claude, ChatGPT and Cursor.
The company said the server is metadata-aware, allowing agents to work with information about relationships, permissions, workflows, validation rules and governance associated with Salesforce capabilities. This is intended to allow actions to inherit existing enterprise controls rather than requiring business rules to be recreated in prompts or custom code.
Salesforce is also introducing the Data 360 MCP Server, which exposes nearly 200 Data 360 APIs through open standards. The company said authorised agents can use it for tasks including building semantic models, transforming data, generating calculated insights, mapping fields, inspecting identity graphs and creating audience segments.
The expansion also includes Slackbot’s MCP Client, which is generally available and enables Slackbot to connect with Salesforce MCP servers and more than 20 partner applications. This allows users to carry out actions such as updating Salesforce opportunities, retrieving contracts and launching workflows from Slack conversations.
“Data 360 Headless lets our agents reach a unified patient context from any surface — through MCP, without waiting for a purpose-built interface for each one,” said Dr. Paramjit “Romi” Chopra, Founder & CEO, MIMIT Health. “What makes those answers clinically trustworthy is the physician-curated ontology sitting between the data and the agent.”
Brendan Ittelson, Chief Ecosystem Officer at Zoom, said, “By integrating our MCP server with Slackbot, we’re making it effortless for millions of users to access their Zoom meeting intelligence through natural conversation — no context switching required.”
The company is extending Headless 360 across its Marketing, Sales, Service and Commerce offerings, as well as MuleSoft, Informatica and Tableau. Salesforce said these capabilities are intended to make functions such as marketing campaigns, sales workflows, service scheduling and commerce capabilities available for use across external applications and AI-driven experiences.
The announcement also includes more than 100 reusable Agent Skills and Plugins, which package business logic and orchestration into reusable capabilities. Salesforce Multi-framework is now generally available for building React applications with Salesforce authentication, security and data access, while the Headless Experience Layer is in open beta for web, mobile, embedded and conversational experiences.
Salesforce said the broader objective is to evolve its platform from application-centric software toward reusable enterprise capabilities that can be accessed by authorised AI agents, applications and user experiences while retaining existing identity, security and governance controls.