July 10, 2025

We are excited to announce the GA release of Model Context Protocol (MCP) support across MuleSoft’s Anypoint Platform — enabling organizations to harness the full power of AI agents by seamlessly connecting them to enterprise systems and data.

This marks a significant step forward in our vision that every organization will become AI-native. By adding support for MCP — the emerging standard for agent-to-tool communication — we’re making it easy for enterprises to turn their existing APIs and integrations into dynamic, agent-ready actions. The result: faster development, richer context, and more intelligent, secure, and responsive agentic experiences across the business.

With these strategic investments, MuleSoft is empowering customers to move beyond experimentation and into production with AI and agents — all while maintaining the governance, flexibility, and scale they rely on today.

AI agents are quickly becoming an essential part of the enterprise technology stack — powering everything from customer support to supply chain optimization. But for agents to be effective, they need access to the full context of your business: data, logic, and systems. That’s where MCP comes in.

Model Context Protocol (MCP) is an emerging standard that enables agents to securely and efficiently interact with enterprise systems. With MuleSoft MCP Support, organizations can now:

• Expose any MuleSoft-managed API or integration as an MCP server
• Enable agents to discover and invoke enterprise actions through a standardized protocol 
• Eliminate the need for writing custom agent-specific code to bring in critical business data

By turning your existing APIs and integrations into MCP-compatible assets, you empower agents to operate with real-time business context — increasing accuracy, reducing hallucinations, and unlocking entirely new levels of workflow automation.

MuleSoft’s implementation of MCP is designed for flexibility, speed, and scale. With just a few clicks, developers can transform their APIs and Mule apps into MCP servers — no complex setup or manual configuration required.

Here’s what’s included:

• Connector-based MCP implementation for APIs and integrations: Make your existing Mule applications agent-ready with minimal effort
• Support for internal, custom, and third-party systems: From NetSuite and SAP to internal tools and databases, expose any system to your AI agents
• Secure, scalable connectivity: Ensure agents only access the systems and actions they’re authorized to use — all managed through the Anypoint Platform

Check out this guide to get started with MCP in Anypoint Platform today

With MCP support, the possibilities are wide-ranging. Here are just a few ways organizations can use MuleSoft to power more intelligent AI agent experiences:

Inventory management agents with full context

A company wants an inventory agent that can consolidate stock levels from Salesforce, NetSuite, and a custom on-prem database. Without MCP, the agent would only access SaaS APIs and operate with incomplete information. With MuleSoft MCP Support, all systems are surfaced through a single, secure layer — ensuring smarter restocking recommendations and fewer fulfillment delays.

Customer support agents with true personalization

An customer agent built on Agentforce can answer billing questions, schedule deliveries, and even create service tickets — but only if it can securely access internal APIs. With MCP-enabled integrations, the agent can execute actionability across the business, delivering highly personalized service that goes beyond the data it has been trained on.

Scalable, secure agentic automation across departments

IT teams can define once and reuse across many agents — ensuring consistency, scalability, and centralized governance. As organizations scale AI agent usage across teams, MuleSoft MCP Support ensures there’s no need to rebuild integrations from scratch for each use case, and that every backend system interaction is secure.

By bringing MCP support into the MuleSoft Anypoint Platform, we’re continuing to deliver on our mission to help organizations connect everything, automate anything — and now, empower agents everywhere.

MCP represents a unified, open approach to agentic integration. Rather than building custom adapters or dealing with fragmented tooling, developers can now use MuleSoft to expose business capabilities in a standardized, agent-ready format.

MuleSoft MCP Support is currently available in Anypoint Exchange.

We've also introduced MuleSoft Governance for Agent Interactions, a set of capabilities to ensure that you can govern and control all your agent interactions with your applications and other agents. As part of this release, MuleSoft Flex Gateway now supports MCP and Agent2Agent (A2A) protocols -- enabling you to apply API policies to these endpoints and monitor agent interactions. Learn more.       

Along with the GA status and support, this connector version includes new features, performance improvements, and bug fixes over the previous Beta version. Learn more.

New innovations around agentic communication are planned have continued to release across the platform, including: 

• Exposing the generative AI capabilities in our developer tooling as MCP servers to be used by MCP clients, like Cursor and Windsurf
• Support for additional agent frameworks, like the Agent2Agent (A2A) Protocol
• Enhancements to MuleSoft for Agentforce: Topic Center to provide monitoring capabilities across third party systems being used by agents

As the role of AI agents in the enterprise continues to evolve, we’re committed to giving our customers the tools they need to lead in this new era.

Learn more about how MuleSoft's MCP Support can transform your organization's AI capabilities.