mcp-registry/mcp-server-azure-devops

    ==================
      
       /// MCP ///
      /// MCP ///
        
    ==================
        
    [server:online]
    [protocol:ready]

mcp-server-azure-devops

by Tiberriver256

Model Context Protocol (MCP) server that exposes Azure DevOps resources (projects, repos, work-items, pipelines, etc.) to AI assistants through a standardised tool interface.

279
79
Specialized
GitHubDocumentationNPM

Installation

1. Prerequisites
• Node.js ≥ 18
• Git
• Azure DevOps Personal-Access-Token (PAT) with at least **Code (read/write)**, **Build (read)** and **Work Items (read/write)** scopes.
2. Clone & install
git clone https://github.com/Tiberriver256/mcp-server-azure-devops.git cd mcp-server-azure-devops npm install # or: pnpm i
3. Build (TypeScript → JS)
npm run build # runs tsc
4. Configure the server
Create a `.env` file (or set env-vars some other way):
PORT=3080 # Port the server will listen on AZURE_DEVOPS_ORG=myOrg # Azure DevOps organisation name AZURE_DEVOPS_PROJECT=myProj # Default project (optional) AZURE_DEVOPS_PAT=xxxxxxxxxx # Your PAT token MCP_OPENAI_API_KEY=sk-... # Optional – for OpenAI relay
You can also pass the variables directly when starting the process.
5. Start in production
npm run start # compiled build in dist/
or hot-reload during development:
npm run dev # ts-node + nodemon
6. Docker (optional)
docker build -t mcp-azure . docker run -p 3080:3080 -e AZURE_DEVOPS_ORG=myOrg -e AZURE_DEVOPS_PAT=xxxxx mcp-azure
7. Reverse-proxy / TLS
Put the service behind nginx / Caddy if you need HTTPS and rate-limiting.
8. Verify
curl http://localhost:3080/v1/health # → { "status": "ok" }

Documentation

License: MIT License
Updated 7/30/2025

Table of Contents