Azure DevOps MCP Server

What is Azure DevOps MCP Server?

This MCP (Model Context Protocol) server provides tools for interacting with Azure DevOps services through AI assistants. It uses an entity-based architecture that groups operations by resource type (projects, repositories, work items, pull requests, pipelines) rather than exposing many atomic tools. It is intended for developers and teams who want to manage Azure DevOps resources via AI-powered chat interfaces.

How to use Azure DevOps MCP Server?

Install the package, configure it via environment variables or a config/azuredevops.json file, then run the server (node build/index.js or via Docker). Required environment variables are ADO_ORGANIZATION and ADO_PAT. The server exposes entity tools that accept an operation field (e.g., list, get, create) along with parameters specific to that operation.

Key features of Azure DevOps MCP Server

• Entity-based organization: projects, repositories, work items, pull requests, pipelines.
• Cursor-based pagination with continuation tokens for list operations.
• Comprehensive error handling with categorized errors and troubleshooting tips.
• Consistent interface across all entity tools for operations and parameters.
• Configuration via environment variables or a JSON configuration file.

Use cases of Azure DevOps MCP Server

• List all projects in an organization and retrieve details of a specific project.
• List Git repositories in a project and explore branches within a repo.
• Get details of a work item or create new work items (e.g., tasks, user stories).
• List pull requests in a repository with filtering by status and pagination.
• List and inspect CI/CD pipelines in a project.

FAQ from Azure DevOps MCP Server

What are the required configuration values?

You must provide ADO_ORGANIZATION (Azure DevOps organization name) and ADO_PAT (Personal Access Token). The default project (ADO_PROJECT) is optional.

How does pagination work?

All list operations support cursor-based pagination. You can set maxResults to control page size and use a continuationToken from a previous response to fetch the next page.

What entity tools are available?

The server provides entity tools for Projects, Repositories, Work Items, Pull Requests, and Pipelines. Each tool offers multiple operations (e.g., list, get, create).

How is authentication handled?

Authentication uses a Personal Access Token (PAT) provided either via the ADO_PAT environment variable or the credentials.pat field in the configuration file.

What runtime is required?

The server runs on Node.js. It is built with npm run build and executed with node build/index.js. A Docker image is also available.