Singapore – Salesforce has announced a significant expansion of Headless 360, its platform architecture designed to let AI agents securely discover and act across enterprise systems without requiring custom integrations for every new use case.
The update arrives as organisations increasingly deploy AI agents as a primary interface for enterprise software, yet struggle to let those agents act securely across business systems.
The expanded rollout introduces new Model Context Protocol (MCP) servers, Data 360 capabilities, Slack integrations, developer tools, and more than 100 reusable Skills, alongside headless experiences spanning every major Salesforce cloud.
In effect, the update is designed to let any authorised AI agent discover, understand, and act across an enterprise’s Salesforce environment.
Rather than functioning as a set of standalone applications, Headless 360 reframes the Salesforce platform as a collection of reusable enterprise capabilities.
Consequently, organisations can reuse business capabilities they have already built on Salesforce instead of rebuilding integrations each time a new AI experience is introduced.
Salesforce says this approach preserves existing identity, permissions, metadata, workflows, governance, and business logic across an enterprise’s systems.
The company frames Headless 360 as the architectural foundation for the next phase of its platform, extending every Salesforce cloud beyond the confines of a single application.
Salesforce has positioned the move as a continuation of its history with open enterprise technology, noting that it introduced the enterprise web API more than 27 years ago.
That earlier development made enterprise data and business functionality accessible beyond a single application, and Salesforce says Headless 360 applies the same principle to the agentic era.
One early example cited by the company is Engine, which reportedly launched its AI support agent, EVA, within 12 days.
Salesforce says EVA now resolves half of all customer chat interactions without human intervention.
Turning to the substance of the announcement, Salesforce has expanded Headless 360 across several layers of what it calls the Agentic Enterprise.
Among the additions is the Headless 360 MCP Server, which brings Salesforce capabilities to AI agents through open standards.
A separate Data 360 MCP Server extends customer context beyond the Salesforce environment itself.
Meanwhile, a new Slackbot MCP Client is intended to bring enterprise actions directly into workplace conversations.
Salesforce has also released Agent Skills and Plugins, comprising more than 100 reusable Skills that package business logic into governed capabilities.
A Headless Experience Layer, now in open beta, is designed to help developers build web, mobile, embedded, and conversational experiences.
In addition, a Salesforce Multi-framework tool allows developers to build React applications with native Salesforce authentication, security, and governance.
Beyond these core releases, Salesforce has expanded Headless 360 capabilities across Marketing, Sales, Service, Commerce, MuleSoft, Informatica, Tableau, and the wider platform.
Turning to security, Salesforce argues that as organisations deploy more agents, open APIs alone are insufficient.
Instead, the company says enterprises require a governed foundation that allows every agent to discover, understand, and safely act across business systems.
At the centre of that foundation sits the Headless 360 MCP Server, built on the open MCP standard.
Rather than requiring developers to manually specify every object, API, workflow, or business rule an agent might need, the server is designed to let agents dynamically discover and invoke Salesforce capabilities in real time.
Salesforce says this functionality extends to agents operating within Agentforce, Claude, ChatGPT, Cursor, and other AI platforms.
The company draws a distinction here, stating that most MCP servers expose APIs, whereas its server exposes trusted business capabilities.
Because the server is metadata-aware, Salesforce says agents do not simply discover endpoints but also understand the relationships, permissions, workflows, validation rules, and governance underpinning how a business actually operates.
As a result, every action reportedly inherits the trusted context an organisation has already established within Salesforce, allowing agents to execute work without developers recreating enterprise logic in prompts or custom code.
Salesforce describes the outcome as a secure foundation for the Agentic Enterprise, one that lets organisations extend Salesforce beyond traditional applications while retaining existing identity, security, governance, and business logic.
Turning to data, Salesforce notes that enterprise AI is only as effective as the context supporting it.
With that in mind, the company has introduced the Data 360 MCP Server, which extends customer context beyond the Salesforce user interface and exposes close to 200 Data 360 APIs through open standards.
Previously, headless experiences were largely limited to querying customer data.
With the new server, Salesforce says authorised agents can build semantic models, transform data, generate calculated insights, map fields, inspect identity graphs, create audience segments, activate campaigns, and stand up new data streams through natural language.
Every Data 360 MCP Server will also include prebuilt Skills covering common workflows such as code extensions, transformations, and activation.
Salesforce says these Skills are intended to move customers from a single prompt to govern execution with minimal setup, and are expected to become generally available later this month.
Agentforce Coworker, built on Data 360, is cited by Salesforce as an example of Headless 360 in practice.
The tool is designed to bring customer context into daily workflows, enabling employees to reason, orchestrate, and act across Salesforce, Slack, Gmail, and other enterprise applications from a single conversational interface.
Commenting on the healthcare application of Data 360 Headless, Dr Paramjit “Romi” Chopra, Founder and Chief Executive of MIMIT Health, elaborated, “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. What makes those answers clinically trustworthy is the physician-curated ontology sitting between the data and the agent. That’s where intelligence lives, and it’s why we can hold onto one question, one answer, every time.”
Turning to workplace collaboration, Salesforce notes that Headless 360’s value grows when agents operate within tools employees already use, including Slack.
To that end, Slackbot’s MCP Client is now generally available, allowing Slackbot to connect securely with the Salesforce platform, all Salesforce MCP servers, and more than 20 partner applications.
Those partner applications include Atlassian, Box, Canva, Docusign, Notion, and Zoom.
As a result, employees can reportedly update Salesforce opportunities, retrieve contracts, launch workflows, and coordinate work across enterprise systems directly from Slack conversations, while inheriting existing permissions, identity, and governance policies.
Speaking on the Zoom integration, Brendan Ittelson, Chief Ecosystem Officer at Zoom, said, “Zoom’s open platform strategy is built on meeting customers where they work. 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. This is what the agentic future of work looks like.”
Looking beyond individual tools, Salesforce frames Headless 360 as representative of a broader shift in how its platform is evolving for the agentic era.
For decades, the company notes, enterprise software has been organised around discrete applications.
Headless 360, by contrast, is intended to transform every Salesforce cloud into a set of reusable enterprise capabilities that AI agents, applications, and experiences can consume.
Within Marketing, the now generally available Marketing Engagement MCP Server allows marketers to manage journeys, campaigns, data extensions, and content through natural language across any interface.
Within Sales and revenue, Agentforce Sales and Self-Service Buying are designed to extend revenue workflows beyond the CRM, spanning applications, storefronts, and AI-powered buying experiences.
Within Service, Headless Service extends dispatching, Field Service scheduling, and customer support workflows into external applications, messaging platforms, and conversational tools.
Within Commerce, Headless Commerce is intended to enable AI-powered storefronts while giving developer agents access to the commerce capabilities needed to build and launch experiences quickly.
On Trusted data and integration, MuleSoft and Informatica expose integration tools, APIs, data quality, governance, metadata, and enterprise context as reusable MCP services, which Salesforce says reduces the need for custom integration work.
More broadly, Headless 360 is designed to operate across the wider enterprise ecosystem, expanding trusted context through zero-copy data partnerships.
It also gives customers access to a growing set of capabilities through AgentExchange, and connects Salesforce actions to external AI platforms including those from Anthropic, OpenAI, Google, and Amazon Web Services.
Turning to developer tooling, Salesforce notes that making its platform available everywhere required more than exposing APIs alone. It required a complete developer platform.
The announcement also expands the tools developers use to build agent-powered experiences:
- Agent Skills and Plugins, with more than 100 reusable Skills that package orchestration, validation, resources, and business logic into governed capabilities
- Salesforce Multi-framework, now generally available, enabling React applications with native Salesforce authentication, security, and data access
- Headless Experience Layer, now in open beta, helping developers build web, mobile, embedded, and conversational experiences powered by Salesforce
Together, Salesforce says these tools support the full lifecycle of enterprise AI development, from discovering capabilities to orchestrating business logic to shipping production-ready applications.
With Headless 360, the company says it is continuing that shift by turning trusted business capabilities into reusable services that any authorised agent, application, or experience can access.
