logo
Free, unlimited AI code reviews that run on commit
git-lrc git-lrc GitHub Install Now We'd appreciate a star git-lrc - Free, unlimited AI code reviews that run on commit | Product Hunt git-lrc - Free, unlimited AI code reviews that run on commit | Product Hunt

AzureDevOps-MCP-Adapter

A standardized Model Context Protocol (MCP) server providing comprehensive programmatic access to core Azure DevOps features, including item management, source control operations, pipeline visibility, and project lifecycle administration. This utility streamlines development lifecycle automation.

Author

AzureDevOps-MCP-Adapter logo

RyanCardin15

MIT License

Quick Info

GitHub GitHub Stars 47
NPM Weekly Downloads 0
Tools 1
Last Updated 2026-02-19

Tags

azuredevopsdevopsazureazure devopsazuredevops mcpryancardin15 azuredevops

Azure DevOps Management Context Protocol Server

Star History Chart

This robust adapter furnishes a standardized interface via the Model Context Protocol (MCP) for interacting with Microsoft Azure DevOps services. It enables external systems and AI agents to orchestrate work items, manage source repositories, inspect build artifacts, and control project configurations.

Operational Summary

This server abstracts the underlying REST APIs of Azure DevOps, presenting a unified, protocol-compliant endpoint for managing standard ALM entities across cloud and on-premises installations.

Core Functionality Domains

The integration exposes capabilities categorized across eight primary functional areas:

Work Item Orchestration

  • Execution of WIQL queries for item retrieval.
  • Fetching granular data for specific work item identifiers.
  • Advanced searching across the work item store.
  • Retrieval of recently modified or user-assigned tasks.
  • Creation, updating, state transitions, and reassignment of tickets.
  • Managing hierarchical links and bulk modifications.

Board and Iteration Management

  • Querying team-specific Kanban boards and column layouts.
  • Retrieving and manipulating work items located on boards.
  • Accessing sprint definitions, current status, and team capacity planning.
  • Enumerating team membership.

Project Configuration Tools

  • Listing and creation of top-level projects.
  • Discovery of project metadata, area paths, and iteration structures.
  • Management of process templates and available work item schemas (fields).

Source Code (Git) Operations

  • Repository enumeration and detailed metadata retrieval.
  • Branch listing and content browsing (file retrieval).
  • Inspection of commit lineage.
  • Management of Pull Requests (listing, creation, detail fetching, review approval, and merging).

Quality Assurance & Testing Harness

  • Triggering automated test suites.
  • Reporting on test execution metrics and agent configuration.
  • Analysis tools: test impact assessment, flakiness diagnostics, and gap identification.
  • Management of exploratory testing sessions and result conversion into actionable work items.

Security and Compliance (DevSecOps)

  • Initiation and result retrieval for security scans (SAST/DAST).
  • Tracking reported vulnerabilities and generating compliance audit reports.
  • Policy enforcement checks and security awareness tracking.
  • Secret management utilities (rotation, auditing, vault integration).

Package and Artifact Handling

  • Management of artifact feeds and package version lifecycles (publishing, promotion, deletion).
  • Container image registry operations (listing, tagging, scanning).
  • Universal package handling and dependency reporting.

AI-Assisted Development Enhancements

  • Tools for generating AI-driven code critiques and suggestions (smells, optimizations).
  • Predictive analytics for effort estimation and bug probability.
  • Metrics retrieval for developer velocity and code quality trends.
  • Intelligent refinement of pending work items and automation opportunity detection.

Deployment and Initialization

Rapid Deployment via NPX

Execution is straightforward using the Node Package Execute utility:

bash npx @ryancardin/azuredevops-mcp-server@latest

Client Integration (Cursor Example)

Direct installation into supported IDE/MCP clients is facilitated via specialized deep links that pre-populate configuration details:

Add Azure DevOps MCP to Cursor

Crucial Step: Following automated setup, environment variables within the client's MCP settings must be updated with genuine organizational access details.

Server Configuration Variables

The adapter relies on environment variables for authentication and targeting. Configuration schemas differ slightly for cloud vs. on-premises deployments.

Variable Name Purpose Cloud Required On-Prem Required
AZURE_DEVOPS_ORG_URL Root URL of the Azure DevOps instance Yes Yes
AZURE_DEVOPS_PROJECT Default project context Yes Yes
AZURE_DEVOPS_IS_ON_PREMISES Boolean flag (true/false) No Yes (if applicable)
AZURE_DEVOPS_AUTH_TYPE Method: pat, ntlm, basic, entra No No
AZURE_DEVOPS_PERSONAL_ACCESS_TOKEN Security token for PAT authentication Conditional Conditional
AZURE_DEVOPS_USERNAME/PASSWORD Credentials for legacy auth methods No No
ALLOWED_TOOLS Restrict exposed functionality via comma-separated list No No

For on-premises setups, additional parameters such as AZURE_DEVOPS_COLLECTION and AZURE_DEVOPS_API_VERSION are mandatory.

Interaction Examples

Client interactions are structured as JSON messages directed to the MCP endpoint.

**Querying Active Work Items (WIQL): **

{ "tool": "listWorkItems", "params": { "query": "SELECT [System.Id], [System.Title] FROM WorkItems WHERE [System.State] = 'Active' ORDER BY [System.CreatedDate] DESC" } }

**Submitting a New Code Review Approval: **

{ "tool": "approvePullRequests", "params": { "repositoryId": "repo-guid-123", "pullRequestId": 42, "reviewerId": "user-alias" } }

Technical Structure

The codebase divides responsibilities into distinct layers for clarity and maintainability:

  1. Service Layer: Handles all direct, low-level communication with the Azure DevOps REST APIs (e.g., WorkItemService, GitService).
  2. Tools Layer: Wraps the service calls, applying necessary input validation and formatting responses specifically for the MCP protocol contract (e.g., WorkItemTools).

Tool registration occurs centrally within the main entry file (index.ts), where each exposed function is mapped to its corresponding implementation, schema definition, and documentation.

Security Note: PAT Generation

When configuring a Personal Access Token (PAT) for Azure DevOps Services, ensure the token's scope grants at minimum Read/Write access to Work Items and Code, alongside Read access for Project and Build/Release data, as required by the activated tools.

See Also

`