mcpField
D
d365-agent-mcpserver-ts\n\nTypeScript implementation of the MCP Server for the Dynamics 365 AI Agent.

d365-agent-mcpserver-ts\n\nTypeScript implementation of the MCP Server for the Dynamics 365 AI Agent.

Local
github.com/ntrtdAI & Agents0Updated 76w ago
Repository

About d365-agent-mcpserver-ts\n\nTypeScript implementation of the MCP Server for the Dynamics 365 AI Agent.

d365-agent-mcpserver-ts is a TypeScript/Node.js implementation for defining and exposing tools to MCP clients. It can serve custom backend or computational logic; its README notes that its Dynamics 365 connectivity is limited by a related client’s known issues.

Key features

  1. 01Define tools with input and output schemas
  2. 02Expose custom tool implementations to MCP clients
  3. 03Support tools for backends, APIs, or custom logic
  4. 04Designed for Docker packaging

Use cases

  1. 01Expose custom logic to an MCP client
  2. 02Provide tools for a backend system or API
  3. 03Package an MCP tool server as a Docker container

MCP support

MCP clients can discover the server's tools and schemas, then call tools and receive results or errors. The server implements tool definitions and execution.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Dynamics 365 AI Agent - MCP Server (TypeScript) (d365-agent-mcpserver-ts)

This repository contains the source code for a TypeScript/Node.js implementation of a Model Context Protocol (MCP) Server. This server can be used to expose various tools and functionalities to MCP-compliant clients.

Overview

The d365-agent-mcpserver-ts is designed to act as a server that AI agents (like those built with LangGraph in d365-agent-orchestrator) can connect to via an MCP client (such as d365-agent-mcpclient-ts). It can expose tools related to Dynamics 365 or other backend systems.

Key Features & Responsibilities

  • MCP Server Implementation: Implements the server-side requirements of the Model Context Protocol using the TypeScript MCP SDK (from submodules/typescript-sdk/packages/mcp-server or a corresponding npm package).
  • Tool Definition & Exposure: Allows developers to define and expose custom tools, specifying their input/output schemas and implementation logic.
  • D365 Integration (Potential/Conditional):
    • If used for Dynamics 365 integration, it would ideally consume a TypeScript D365 OData client generated by d365-agent-odataclient-ts.
    • Current Status Note: The d365-agent-odataclient-ts has known issues for D365 connectivity. Therefore, for D365 specific tools, the d365-agent-mcpserver-dotnet (which uses a working .NET OData client) is the current primary and recommended MCP server in the D365 AI Agent architecture.
  • Alternative Uses: This TypeScript MCP server can be valuable for:
    • Exposing tools that interact with other (non-D365) backend systems, databases, or APIs.
    • Implementing custom logic or computational tools that don't require D365 OData access.
    • Serving as a lightweight MCP tool server if the d365-agent-odataclient-ts issues are resolved in the future.
  • Containerization: Designed to be packaged as a Docker container for deployment.

Technology Stack

  • TypeScript / Node.js
  • Web framework for HTTP transport (e.g., Express, Fastify, or built into the MCP Server SDK).
  • D365 Agent TypeScript MCP SDK (from submodules/typescript-sdk or its npm package).
  • Potentially d365-agent-odataclient-ts if used for D365 and issues are resolved.

Interaction Flow (General MCP Server)

  1. An MCP client (e.g., d365-agent-mcpclient-ts from d365-agent-orchestrator) connects to this server.
  2. The client can use discoverTools to get a list of available tools and their schemas.
  3. The client sends an MCP CallTool request for a specific tool with parameters.
  4. The d365-agent-mcpserver-ts executes the tool's implementation logic.
  5. The result or error is packaged into an McpResponse and sent back to the client.

Getting Started

(Details to be added - typically involves cloning, Node.js/TypeScript setup, installing dependencies, defining tools, and running the server). Refer to the implementation.md in this repository for a detailed development plan.

Contribution

(Details on contribution guidelines if applicable).

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.