The OpenEdge MCP Server sits between the AI ecosystem and PAS for OpenEdge services, acting as a secure bridge that exposes OpenEdge APIs to AI workflows.

It supports an AI ecosystem that includes assistants, chatbots, tools, workflows, and AI applications. AI assistants and agents typically connect through an MCP Client, which communicates with the MCP server. Through this integration, the OpenEdge MCP Server delivers intelligent automation and conversational experiences.

The OpenEdge MCP Server consists of four main layers that ensure secure, scalable, and observable operations:
  • Security gateway for dual authentication, access control, and rate limiting.
  • API converter for OpenAPI parsing, tool generation, and route management.
  • Management layer featuring CLI tools, Docker runtime, and automated infrastructure.
  • Observability layer for structured logging, audit events, and request timing.

The OpenEdge MCP Server connects to PAS for OpenEdge services through REST and WEB interfaces, enabling seamless communication between OpenEdge applications and MCP components.

It exposes essential OpenEdge components as secure, discoverable tools for AI workflows. These tools provide a structured interface that enables natural language-driven operations. The following OpenEdge components are exposed:
  • ABL business logic from Progress Application Server (PAS) for OpenEdge.
  • REST and Web APIs defined in OpenAPI specifications.
  • OpenEdge Business Entities as discoverable and callable tools.
Each MCP tool includes attributes that maintain security, enforce compliance, and support reliable execution, such as:
  • Descriptions, parameter schemas, and entity tags for context-aware execution.
  • Scope-based access control to enforce least privilege.
  • Safe execution and predictable responses.
  • Capturing invocation details for compliance and governance.