MCP Server Azure DevOps

An MCP server that gives AI assistants full programmatic access to Azure DevOps — enabling natural language management of work items, repositories, branches, pull requests, pipelines, wikis, and code search across both Azure DevOps Services (cloud) and on-premises Server.

Evaluated Mar 06, 2026 (0d ago) vlatest
Homepage ↗ Repo ↗ Developer Tools azure-devops work-items pipelines pull-requests repos git ado microsoft
⚙ Agent Friendliness
77
/ 100
Can an agent use this?
🔒 Security
86
/ 100
Is it safe for agents?
⚡ Reliability
81
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
78
Documentation
80
Error Messages
68
Auth Simplicity
68
Rate Limits
72

🔒 Security

TLS Enforcement
100
Auth Strength
85
Scope Granularity
82
Dep. Hygiene
80
Secret Handling
80

Azure DevOps PAT with scoped permissions. Use minimal PAT scopes — read-only if only querying. Organization-level PAT has broad access. Prefer project-scoped tokens.

⚡ Reliability

Uptime/SLA
88
Version Stability
80
Breaking Changes
78
Error Recovery
78
AF Security Reliability

Best When

Your engineering team uses Azure DevOps as the primary ALM and you want an AI assistant to reduce context-switching between the ADO UI and your coding environment.

Avoid When

Your source control and issue tracking are on GitHub or GitLab — ADO-specific tooling won't help there.

Use Cases

  • • Create, update, and triage work items (bugs, tasks, user stories) using natural language
  • • Browse repository files, create branches, and commit changes through an AI assistant
  • • Open pull requests and add review comments programmatically
  • • Trigger and monitor pipeline runs and retrieve build artifacts
  • • Search code and work items across multiple ADO projects from a single query

Not For

  • • GitHub-hosted repositories — use a GitHub MCP server instead
  • • Non-Microsoft DevOps platforms (GitLab, Jira, Bitbucket)
  • • Organizations without Azure DevOps access or licenses

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
Yes ↗
SDK
No
Webhooks
No

Authentication

Methods: personal_access_token azure_identity azure_cli
OAuth: No Scopes: No

Three methods: PAT (simplest), Azure Identity / DefaultAzureCredential (for managed identity / service principal), or Azure CLI session reuse. Both Azure DevOps Services and on-premises Server are supported.

Pricing

Model: open_source
Free tier: Yes
Requires CC: No

MIT licensed. Requires an Azure DevOps organization (free tier available from Microsoft for up to 5 users; paid plans for larger teams).

Agent Metadata

Pagination
offset
Idempotent
Partial
Retry Guidance
Not documented

Known Gotchas

  • ⚠ PAT tokens have configurable expiration — long-running agent deployments need token rotation strategy
  • ⚠ ADO REST API rate limits (varies by tier) can throttle agents making rapid sequential calls
  • ⚠ On-premises ADO Server may have different API versions than cloud Services — verify compatibility
  • ⚠ Work item creation is not idempotent; agents must check for existing items before creating to avoid duplicates
  • ⚠ Azure Identity auth requires correct role assignments in AAD — setup can be complex in enterprise environments

Alternatives

mcp-github mcp-gitlab mcp-jira

Full Evaluation Report

Detailed scoring breakdown, competitive positioning, security analysis, and improvement recommendations for MCP Server Azure DevOps.

$99
API endpoint ↗ Agent guide ↗ Report inaccuracy

Scores are editorial opinions as of 2026-03-06.

5229
Packages Evaluated
26151
Need Evaluation
173
Need Re-evaluation
♥
Community Powered