azure-devops
Exposes Azure DevOps functionality via MCP: Boards, Repos, Pipelines, Artifacts, Test Plans, and Wiki tools for AI agents.
claude mcp add --transport stdio jordiag-azure-devops-mcp-server dotnet path/to/Dotnet.AzureDevOps.Mcp.Server.dll \ --env AZDO_PAT="your-personal-access-token-or-placeholder" \ --env AZDO_ORG_URL="https://dev.azure.com/your-org" \ --env ASPNETCORE_URLS="http://0.0.0.0:5000"
How to use
This MCP server exposes a set of Azure DevOps operations through the Model Context Protocol (MCP). It wraps multiple Azure DevOps areas—Boards, Repos, Pipelines, Artifacts, Test Plans, and more—into a uniform MCP endpoint at /mcp, accessible via Server-Sent Events (SSE). AI agents can discover available tools at runtime and invoke them with structured function calls, enabling automation for creating work items, managing PRs, queuing builds, handling artifacts, and other DevOps tasks without bespoke SDK calls.
To use it, run the server (for example with dotnet) and connect your MCP client to the hosted /mcp endpoint. The server hosts multiple tool namespaces under the MCP surface, each corresponding to an Azure DevOps domain (Boards, Repos, Pipelines, Artifacts, Test Plans, Project Settings, and Search). You can discover available tools at runtime and execute operations such as creating a work item, listing pull requests, queuing a build, or querying work item histories. The design aims to provide a stable, high-level interface while delegating the heavy lifting to the underlying Azure DevOps SDKs and REST APIs behind the scenes.
How to install
Prerequisites:
- .NET 10 SDK installed on your machine or host environment
- Git
- Access to an Azure DevOps organization (organization URL and PAT)
- Optional: Docker if you prefer containerized runs
Installation steps (manual install):
-
Clone the repository: git clone https://github.com/J Jordiag/azure-devops-mcp-server.git cd azure-devops-mcp-server
-
Restore and build the solution: dotnet restore dotnet build -c Release
-
Run the MCP server directly (example): dotnet path/to/Dotnet.AzureDevOps.Mcp.Server.dll
Ensure environment variables are set as needed (see mcp_config)
-
Optional: run via Docker
- Build image (Dockerfile should expose the /mcp endpoint): docker build -t azure-devops-mcp-server:latest .
- Run container: docker run -e ASPNETCORE_URLS=http://0.0.0.0:5000 -p 5000:5000 azure-devops-mcp-server:latest
Using with NPX or Node is not applicable here since this is a .NET-based MCP server. If you already use a .NET build pipeline, you can integrate the server startup as part of your CI/CD workflow.
Additional notes
Tips and considerations:
- The server focuses on Azure DevOps operations wrapped as MCP tools; ensure your Azure DevOps organization URL and authentication token (PAT) are correctly configured in the environment variables (AZDO_ORG_URL, AZDO_PAT).
- The MCP endpoint is served over SSE at /mcp; your MCP client should support SSE streaming to fetch tool definitions and invoke them.
- This is a preview release; API surfaces may change. Regularly pull updates from the repository to stay aligned with the latest MCP tool definitions.
- If you run behind a proxy or in a restricted network, configure ASPNETCORE_URLS and any necessary HTTP proxy settings.
- For production deployments, consider securing the endpoint (HTTPS), rate limiting, and proper secret management for PATs.
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
mobile-dev
This is a MCP designed to manage and interact with mobile devices and simulators.
zerodha
Zerodha MCP Server & Client - AI Agent (w/Agno & w/Google ADK)
symfony
A Symfony package designed for building secure servers based on the Model Context Protocol, utilizing Server-Sent Events (SSE) and/or StreamableHTTP for real-time communication. It offers a scalable tool system tailored for enterprise-grade applications.
mcp-document-converter
MCP Document Converter - A powerful MCP tool for converting documents between multiple formats, enabling AI agents to easily transform documents.
mcp -java-sse-http-demo
Java Model Context Protocol SSE HTTP Server with Jetty