mcp -azure-devops

This MCP server provides a bridge to Azure DevOps resources through the Model Context Protocol, exposing tools to interact with projects, work items, repositories, pull requests, branches, and pipelines. It enables AI assistants to securely authenticate and perform common DevOps workflows via natural language queries, such as creating or updating work items, retrieving repository contents, or triggering pipelines. The server is designed with a modular, feature-based architecture so you can extend or customize individual capability modules (work-items, projects, repositories, etc.) as needed. Authentication can be configured using PATs, Azure Identity, or Azure CLI, depending on your environment and Azure DevOps setup.

Prerequisites:

• Node.js v16 or newer
• npm or yarn
• Access to a Azure DevOps organization (PAT or Azure Identity credentials)
• Optional: Azure CLI installed for az login scenarios

From a fresh checkout or when running from npm:

1. Install dependencies and build (from source):

npm ci
cp .env.example .env   # then edit values
npm run build
npm start              # runs: node dist/index.js

2. Run in development mode with auto-reload (for iterative development):

npm run dev

3. Running via npx (published package) without cloning:

npx -y @tiberriver256/mcp-server-azure-devops

4. Optional alternative (if you clone and run locally):

# after cloning
npm ci
cp .env.example .env
npm run build
npm start

Environment variables control authentication and default project behavior. Common variables include AZURE_DEVOPS_ORG_URL (organization URL), AZURE_DEVOPS_AUTH_METHOD (pat, azure-identity, or azure-cli), and AZURE_DEVOPS_DEFAULT_PROJECT. For on-prem Azure DevOps Server, use PAT authentication and provide the server URL in AZURE_DEVOPS_ORG_URL. If you plan to run via Claude Desktop or Cursor AI, ensure your config file uses the appropriate command and environment settings as shown in the README examples. If you encounter authentication issues, verify your credentials and that the configured identity method has sufficient permissions to access the requested resources. Logs can be adjusted with LOG_LEVEL in the environment to help diagnose problems.