mcp-azure-devops

This MCP server provides an integration layer that lets AI assistants interact with Azure DevOps by exposing common DevOps actions as natural-language enabled MCP commands. Through the server, you can query and manage work items, retrieve project and team information, and navigate team iterations and area paths. Capabilities include querying work items with WIQL, creating and updating work items, adding and viewing comments, and establishing parent-child relationships between items. You can also fetch projects, list teams, inspect team members, and view team iterations and area paths. Use these capabilities to enable conversational agents to perform DevOps tasks without directly calling the REST API.

To use it, run the MCP server locally (see installation). Once running, you can instruct your AI assistant to perform actions such as querying active bugs, creating a user story, updating a work item with a status change, or listing all projects and their iterations. The server translates your natural-language requests into the appropriate Azure DevOps REST API calls and returns structured results that your assistant can present to the user.

Prerequisites

• Python 3.10+
• Git
• Access to the Azure DevOps REST API with a PAT that has the required scopes

Installation steps

1. Clone the repository git clone https://github.com/Vortiago/mcp-azure-devops.git cd mcp-azure-devops

2. (Optional but recommended) Create a virtual environment python -m venv venv source venv/bin/activate # on macOS/Linux venv\Scripts\activate # on Windows

3. Install in development mode (for local development and testing) uv pip install -e ".[dev]"

4. Install from PyPI (if you prefer a standard install outside of development) pip install mcp-azure-devops

5. Prepare environment variables Create a .env file in the project root or export the variables: AZURE_DEVOPS_PAT=your_personal_access_token AZURE_DEVOPS_ORGANIZATION_URL=https://your-organization.visualstudio.com or https://dev.azure.com/your-organisation

6. Run the MCP server locally (development mode) mcp dev src/mcp_azure_devops/server.py

7. Optional: install for Claude Desktop or other clients mcp install src/mcp_azure_devops/server.py --name "Azure DevOps Assistant"

Environment variables:

• AZURE_DEVOPS_PAT: Personal Access Token with necessary scopes for Azure DevOps API access.
• AZURE_DEVOPS_ORGANIZATION_URL: Full URL to your Azure DevOps organization (e.g., https://dev.azure.com/your-organization).

Configuration tips:

• If you rotate PATs, update AZURE_DEVOPS_PAT accordingly.
• The server uses the Azure DevOps REST APIs; ensure network access to azure-devops.io endpoints is allowed.
• Although the repository mentions an official Azure DevOps MCP server, this project is unmaintained. If you need long-term support, prefer the official Microsoft server at https://github.com/microsoft/azure-devops-mcp.

Common issues:

• Invalid PAT or insufficient scopes: re-create a PAT with the required scopes.
• Incorrect organization URL: ensure the URL matches your organization and uses the correct domain (visualstudio.com or dev.azure.com).
• Development mode vs. production mode: use the appropriate installation mode and mirroring tools as per your environment.