On This Page
> resources > docs > api > MCP Server

MCP Server

Enterprise Enterprise Available on the Enterprise plan. Seats on the self-serve plans cannot use this feature. Compare plans Available on the Enterprise plan

MCP Server

The MCP (Model Context Protocol) server lets AI assistants and AI-powered tools connect to your ProcessMind organization and work with your process data directly. Instead of building a custom integration, you configure your MCP client to use ProcessMind’s server, and the assistant can read processes, models, roles, users and process statistics through standard MCP tools.

What Is MCP?

MCP is an open standard that defines how AI applications (the “client”, such as an AI IDE or assistant) connect to external data sources and tools (the “server”). It works like a USB-C port for AI: one standard connection gives assistants access to many tools and resources. ProcessMind implements an MCP server so you can use your existing AI tooling against your organization’s data without writing and maintaining custom glue code.

What the MCP Server Provides

The server exposes your organization’s process content, surfaced as MCP tools and resources. This includes:

  • Processes and BPMN models
  • Roles, RACI and users
  • Documentation and process statistics
  • AI analysis and answers about your processes
  • Organization and environment metadata

The exact tools available depend on your environment’s configuration. The server focuses on modeling, documentation and analysis workflows; work such as dataset uploads, data mappings, simulations and API-key management is not exposed as MCP tools and still uses the REST API.

Prerequisites

Before connecting, make sure you have:

  1. An Enterprise plan subscription: the MCP server is an Enterprise feature
  2. An API key with access to the environments you want to expose: see API Keys
  3. An MCP-capable client, such as an AI-powered IDE or assistant that supports remote MCP servers

Connecting Your MCP Client

The general setup is the same in every MCP client:

  1. Open your client’s MCP server configuration
  2. Add a remote (HTTP) server using the MCP server URL for your environment
  3. Set the x-api-key header to your API key
  4. Save the configuration and restart or reconnect your client
  5. Confirm the connection by asking your assistant to list your processes

A typical configuration entry looks like this:

{
	"mcpServers": {
		"processmind": {
			"url": "<mcp-server-url>",
			"headers": {
				"x-api-key": "<your-api-key>"
			}
		}
	}
}

Use the MCP server URL for your environment and your own API key in the x-api-key header. (OAuth client-credentials tokens are also supported through the token endpoint.)

Security and Best Practices

  • Keep your API key secret: never commit it to source control or expose it in client-side code
  • Scope keys per environment: create dedicated keys for MCP access so you can revoke them independently
  • Review the audit log: MCP-driven actions are performed by your API key and appear in the Audit Log

Related Documentation