MCP Tools

View all workflows and flow services in the current project that are exposed as MCP tools. AI agents can use MCP tools to access integration capabilities. webMethods Hybrid Integration enables AI agents and tools such as AI-enabled IDEs to access workflows and services through the Model Context Protocol.

Model Context Protocol in webMethods Integration

The Model Context Protocol (MCP) is an open protocol that standardizes the way large language models integrate and share data with other applications, tools, and external data sources. MCP also provides a universal interface for reading data and acting on behalf of a user.

Model Context Protocol (MCP) support in webMethods Hybrid Integration provides a standardized way to expose integrations to AI agents. With MCP, you can add existing workflows and flow services as MCP tools and allow AI agents, MCP-enabled IDEs, and other MCP clients to access and run them. webMethods Hybrid Integration implements MCP communication through Streamable HTTP.

When an integration asset is exposed as an MCP tool, AI agents use the tool description that you provide in the webMethods Hybrid Integration user interface to respond to natural language prompts and search for matching workflows and services. Workflows and flow services are not accessible to AI agents by default. You must explicitly expose them as MCP tools to allow access. You can change this setting at any time, or temporarily disable and re-enable an MCP tool.

You can use any standard MCP client or MCP-enabled IDE to access workflows and flow services in webMethods Hybrid Integration.