MCP -ArcGIS-Pro-AddIn
MCP Server with ArcGIS Pro Add-In
claude mcp add --transport stdio nicogis-mcp-server-arcgis-pro-addin dotnet run --project McpServer/ArcGisMcpServer/ArcGisMcpServer.csproj
How to use
This MCP server exposes ArcGIS Pro functionality through MCP tools so you can query and manipulate ArcGIS Pro state from a client (for example Copilot in Agent Mode). The ArcGIS Pro Add-In runs inside ArcGIS Pro and exposes operations via a Named Pipe. The MCP server hosts a set of tools (for example pro.listLayers and pro.countFeatures) that forward requests to the Add-In bridge over IPC and return structured results. Once the server is running, an MCP client can call these tools by their registered tool names and pass parameters as needed (for example the layer name for counting features).
Typical workflow: start ArcGIS Pro with the Add-In loaded, ensure the Named Pipe server is active, then start the MCP server via Visual Studio Copilot integration or by running the MCP manifest. Use tool names like pro.listLayers to retrieve layer information from the active map or pro.countFeatures with a layer argument to get a feature count. You can extend capabilities by adding more tools that call into the BridgeClient to perform ArcGIS Pro operations and return JSON-serializable results.
How to install
Prerequisites:
- Visual Studio 2022 (version 17.14 or later) with .NET desktop development workload
- .NET 8 SDK installed
- ArcGIS Pro installed on the same machine
- ArcGIS Pro SDK for .NET installed
- ArcGIS Pro Add-In development prerequisites (optional if you’re building locally)
Setup steps:
- Clone the repository containing the MCP server and ArcGIS Pro Add-In: git clone <repository-url>
- Install and configure prerequisites on your machine: ensure Visual Studio 2022, .NET 8 SDK, and ArcGIS Pro with its SDK are installed.
- Open the solution in Visual Studio and restore NuGet packages:
- Build the ArcGisProBridgeAddIn project (ArcGIS Pro Add-In) to generate and load the Add-In into ArcGIS Pro.
- Build the MCP server project (ArcGisMcpServer) to produce the executable.
- Ensure the MCP manifest (.mcp.json) is present in the solution root as shown in the README. The manifest should reference the dotnet run invocation for the ArcGisMcpServer project.
- Start ArcGIS Pro with the Add-In loaded so the Named Pipe bridge is active.
- Run the MCP server from Visual Studio Copilot (or manually) so Copilot can read .mcp.json and start the stdio server.
- Use an MCP client to call tools like pro.listLayers or pro.countFeatures with the appropriate arguments.
Additional notes
Tips and notes:
- The MCP server communicates with ArcGIS Pro via a Named Pipe bridge implemented in the Add-In. Ensure ArcGIS Pro is running and the Add-In is loaded before invoking MCP tools.
- The available tools described in the README include pro.listLayers and pro.countFeatures; you can extend tools by adding new methods to ProTools and corresponding client calls.
- The .mcp.json manifest should be placed at the solution root to let Copilot discover and start the server automatically.
- If you modify tool definitions or add new tools, rebuild the MCP server project and restart the server to pick up changes.
- When debugging IPC calls, confirm the ArcGIS Pro Add-In bridge is listening on the Named Pipe (ArcGisProBridgePipe in the sample).
- If you containerize later, consider replacing the stdio transport with a more robust IPC or REST-based transport if needed for production usage.
- The server name in the manifest is arbitrary for your workflow (here it’s arcgis); ensure it matches your client configuration.
Related MCP Servers
spec-coding
Transform feature ideas into production-ready code through systematic Spec-Driven Development 通过系统化的**规格驱动开发**,将功能想法转化为可投入生产的代码
mcpbi
PowerBI MCP server to give LLM clients (Claude, GH Copilot,etc) context from locally running PowerBI Desktop instances.
NugetMcpServer
.NET MCP server that gives LLMs real NuGet API metadata to reduce hallucinations
azure-devops
Exposes Azure DevOps functionality via MCP: Boards, Repos, Pipelines, Artifacts, Test Plans, and Wiki tools for AI agents.
unity -template
Simple template project for controlling Unity via MCP
mcp -graph-api
Model-context-protocol (MCP) server for the Microsoft Graph API in C#